{ "packId": "ai-for-small-business", "packName": "AI for Small Business", "packVersion": "0.1.2", "icon": "🤖", "shortName": "AI for SMB", "description": "A beginner's practical guide to choosing useful AI workflows, prompting with context, researching carefully, supporting customers, automating safely, protecting data, and running a measured 30-day pilot.", "author": "Flash Feed", "language": "en", "tagsVocabulary": [ "foundations", "prompting", "research", "marketing", "operations", "automation", "privacy", "governance", "implementation" ], "items": [ { "id": "aisb-fact-pattern-machine", "shape": "fact", "title": "A Pattern Machine, Not an Oracle", "body": "A language model generates plausible responses from learned patterns. It can be remarkably capable, but fluent wording is not proof that a claim is true.", "illustration": { "kind": "photo", "imagePrompt": "A person at a laptop looking skeptically at a smooth, polished paragraph on the screen, holding a red pen without yet marking it.", "imageSearchTerm": "skeptical laptop reading document", "alt": "A business owner comparing several draft documents", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Barry_Karr_reading_Skeptical_Inquirer.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-pattern-machine.webp" }, "tags": [ "foundations", "capabilities" ], "uid": "10p9xdjiz2tzj" }, { "id": "aisb-fact-probabilistic", "shape": "fact", "title": "Useful Does Not Mean Deterministic", "body": "The same request can produce different answers. Use each AI output as a draft or recommendation whose reliability depends on the task, context, and review process.", "tags": [ "foundations", "reliability" ], "illustration": { "imageSearchTerm": "multiple draft printouts desk", "imagePrompt": "Several slightly different printed drafts of the same letter fanned out on a desk, each with small variations, no legible text.", "alt": "Useful Does Not Mean Deterministic", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Cuerpos.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-probabilistic.webp" }, "uid": "1we50x11rortbd" }, { "id": "aisb-fact-local-expertise", "shape": "fact", "title": "You Supply the Local Truth", "body": "AI usually does not know your current prices, policies, customers, inventory, or judgment. Your business context is what turns a generic answer into useful work.", "tags": [ "foundations", "context" ], "illustration": { "imageSearchTerm": "shop owner inventory shelf", "imagePrompt": "A small business owner checking a handwritten price list against physical inventory on a shelf in their shop, no legible text.", "alt": "You Supply the Local Truth", "depictable": true, "credit": "Pexels · Kampus Production", "creditUrl": "https://www.pexels.com/photo/man-wearing-apron-working-in-store-with-paints-6023610/", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-local-expertise.webp" }, "uid": "8m7bga1gjykro" }, { "id": "aisb-pair-generation", "shape": "pair", "sideA": { "modality": "text", "value": "Generation" }, "sideB": { "modality": "text", "value": "Creating a new draft from instructions and context" }, "tags": [ "foundations", "capabilities" ], "uid": "1tstcbo1gsf3o4" }, { "id": "aisb-pair-analysis", "shape": "pair", "sideA": { "modality": "text", "value": "Analysis" }, "sideB": { "modality": "text", "value": "Finding patterns or implications in information you provide" }, "tags": [ "foundations", "capabilities" ], "uid": "rnghz4m4rexa" }, { "id": "aisb-def-hallucination", "shape": "definition", "term": { "modality": "text", "value": "Hallucination" }, "definition": { "modality": "text", "value": "A confident-sounding AI output that contains invented, unsupported, or incorrect information." }, "tags": [ "foundations", "reliability" ], "curatedDistractors": [ "Grounded response", "Verified fact", "Source citation" ], "uid": "s1sn1xqck7n" }, { "id": "aisb-mcq-fluency", "shape": "mcq", "prompt": { "modality": "text", "value": "A polished AI answer includes three statistics but no sources. What should you do first?" }, "options": [ { "modality": "text", "value": "Publish it because the prose is confident", "short": "Publish; the prose is confident" }, { "modality": "text", "value": "Replace the numbers with larger estimates", "short": "Swap in larger estimates" }, { "modality": "text", "value": "Verify each statistic against reliable sources", "short": "Verify each stat vs sources" }, { "modality": "text", "value": "Ask the AI whether it is certain", "short": "Ask the AI if it's certain" } ], "correctIndex": 2, "explanation": "Fluency is not evidence. Verify material claims using reliable sources before using them.", "tags": [ "foundations", "verification" ], "uid": "1nyaacnpp930j" }, { "id": "aisb-proc-first-experiment", "shape": "procedure", "goal": "Run a safe first AI experiment", "steps": [ "Choose a low-risk task you already understand", "Remove confidential or personal information", "Give the AI context and a clear desired output", "Compare the result with work you trust", "Record what required correction" ], "notes": "Early experiments should teach you where AI helps and where review is still essential.", "tags": [ "foundations", "practice" ], "uid": "1k7sq0d100f1w9" }, { "id": "aisb-fact-workflow-first", "shape": "fact", "title": "Workflow First", "body": "Start with a recurring business problem, then evaluate tools. Buying an AI subscription before naming the job often creates cost without adoption.", "illustration": { "kind": "photo", "imagePrompt": "Editorial documentary photograph of a local small-business owner planning a real daily operating process with blank paper cards beside a laptop, natural window light, authentic neighborhood business setting, restrained colors, no software wireframes, no user-interface design, no visible logos, no readable text, no futuristic holograms.", "imageSearchTerm": "local business owner operations planning", "alt": "A local business owner planning a daily operating process", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Localh2014.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-workflow-first.webp" }, "tags": [ "foundations", "workflow" ], "uid": "1kh4i2y870n2a" }, { "id": "aisb-fact-best-first-task", "shape": "fact", "title": "The Best First Task", "body": "A strong first task is frequent, time-consuming, easy to check, and low consequence if the first draft is wrong.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of a small-business owner sorting routine office tasks on simple blank cards and selecting one manageable task beside a laptop, practical planning, no readable text, no logos.", "imageSearchTerm": "small business owner prioritizing routine tasks", "alt": "A business owner selecting a manageable routine task", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Image_taken_from_page_124_of_%27La_Terra,_trattato_popolare_di_geografia_universale_per_G._Marinelli_ed_altri_scienziati_italiani,_etc._(With_illustrations_and_maps.)%27_(11159467355).jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-best-first-task.webp" }, "tags": [ "foundations", "workflow" ], "uid": "1951lip1rpkwpx" }, { "id": "aisb-fact-augment-before-automate", "shape": "fact", "title": "Augment Before You Automate", "body": "Use AI alongside the existing process before allowing it to act alone. A parallel run exposes errors, exceptions, and missing instructions.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business worker performing a familiar process while checking a laptop-assisted version side by side, parallel comparison, realistic workplace, no readable screens, no logos.", "imageSearchTerm": "small business worker comparing manual digital process", "alt": "A worker comparing an existing process with a digital assistant", "credit": "Unsplash · Vitaly Gariev · Unsplash License", "creditUrl": "https://unsplash.com/photos/fashion-designer-working-on-her-laptop-and-sipping-coffee-1jQRJt15JFw", "subject": "A woman in a striped top with a tape measure round her neck working at a laptop in a sewing studio, holding a coffee cup; sewing machine, paper pattern sketches and a phone on the table beside her, sketches pinned on the wall", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-augment-before-automate.webp" }, "tags": [ "foundations", "automation" ], "uid": "r7817sae6sh" }, { "id": "aisb-pair-high-fit", "shape": "pair", "sideA": { "modality": "text", "value": "High-fit first workflow" }, "sideB": { "modality": "text", "value": "Frequent, structured, reviewable, and low risk" }, "tags": [ "foundations", "workflow" ], "uid": "4ntx2sbez5le" }, { "id": "aisb-pair-low-fit", "shape": "pair", "sideA": { "modality": "text", "value": "Low-fit first workflow" }, "sideB": { "modality": "text", "value": "Rare, ambiguous, irreversible, or high stakes" }, "tags": [ "foundations", "workflow" ], "uid": "1sxhgp111419w9" }, { "id": "aisb-def-workflow", "shape": "definition", "term": { "modality": "text", "value": "Workflow" }, "definition": { "modality": "text", "value": "A repeatable sequence of inputs, decisions, actions, and outputs used to complete a business task." }, "tags": [ "foundations", "workflow" ], "curatedDistractors": [ "Project", "Policy", "Single task" ], "uid": "dlro6c174zlus" }, { "id": "aisb-mcq-first-workflow", "shape": "mcq", "prompt": { "modality": "text", "value": "Which is the strongest first AI workflow for a neighborhood service business?" }, "options": [ { "modality": "text", "value": "Sign vendor contracts", "short": "Sign vendor contracts" }, { "modality": "text", "value": "Draft weekly social posts from approved offers", "short": "Draft weekly posts from offers" }, { "modality": "text", "value": "Automatically approve refunds with no limit", "short": "Auto-approve refunds, no limit" }, { "modality": "text", "value": "Make final hiring decisions", "short": "Make final hiring decisions" } ], "correctIndex": 1, "explanation": "Drafting social posts is frequent, easy to review, and reversible; the other choices carry much higher consequences.", "tags": [ "foundations", "workflow" ], "uid": "xit08gzngz1a" }, { "id": "aisb-proc-task-inventory", "shape": "procedure", "goal": "Build an AI opportunity list", "steps": [ "List tasks repeated at least weekly", "Estimate time spent on each task", "Mark tasks that follow a recognizable pattern", "Rate the cost of an error", "Choose one frequent, low-risk task to test" ], "notes": "A task inventory keeps adoption grounded in work rather than novelty.", "tags": [ "foundations", "workflow" ], "uid": "1622h9vondgrn" }, { "id": "aisb-fact-prompt-brief", "shape": "fact", "title": "Prompt Like a Brief", "body": "A useful request names the task, supplies relevant context, describes the audience, sets constraints, and defines the expected output.", "illustration": { "kind": "photo", "imagePrompt": "Natural editorial photograph of a business owner preparing a detailed project brief at a laptop with organized reference documents and a notebook, calm focused mood, realistic office lighting, no visible brand logos, no readable screen text, no science-fiction interface.", "imageSearchTerm": "business owner writing project brief laptop", "alt": "A business owner preparing a detailed brief at a laptop", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:California_Hardware_Catalog_Title_Page_1.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-prompt-brief.webp" }, "tags": [ "prompting", "context" ], "uid": "1uoa0hg17fhmn6" }, { "id": "aisb-fact-examples", "shape": "fact", "title": "Examples Beat Adjectives", "body": "A real email, proposal, or response often communicates your voice more clearly than instructions such as 'make it professional.' Remove sensitive details before sharing examples.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead photograph of a business owner using a few anonymized example documents to guide a new draft on a laptop, examples visibly redacted with blank blocks, no readable text, no logos.", "imageSearchTerm": "business owner reviewing example documents laptop", "alt": "A business owner using example documents while drafting", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Bw_logo_1000.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-examples.webp" }, "tags": [ "prompting", "examples" ], "uid": "1xzj5p7djlrlx" }, { "id": "aisb-fact-acceptance-criteria", "shape": "fact", "title": "Define Done", "body": "Acceptance criteria make quality review faster. Specify required facts, length, format, tone, prohibited claims, and what the output must help someone do.", "tags": [ "prompting", "quality" ], "illustration": { "imageSearchTerm": "checklist final checkmark clipboard", "imagePrompt": "A hand checking the final box on a clipboard checklist, marking a piece of work as complete and ready, no legible text.", "alt": "Define Done", "depictable": false, "credit": "Pexels · Jakub Zerdzicki · Pexels License", "creditUrl": "https://www.pexels.com/photo/hand-writing-on-notebook-with-checklist-35719571/", "subject": "a hand with a fountain pen ticking boxes on a handwritten notebook checklist headed 'PROJECT...', seven items, the last one reading 'DONE' with its box still empty", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-acceptance-criteria.webp" }, "uid": "170i5yk1adqoua" }, { "id": "aisb-pair-task", "shape": "pair", "sideA": { "modality": "text", "value": "Task" }, "sideB": { "modality": "text", "value": "The job you want the AI to perform" }, "tags": [ "prompting", "brief" ], "uid": "8l0nkp2ny69f" }, { "id": "aisb-pair-context", "shape": "pair", "sideA": { "modality": "text", "value": "Context" }, "sideB": { "modality": "text", "value": "Business information needed to perform the task well" }, "tags": [ "prompting", "brief" ], "uid": "hr2over5nke4" }, { "id": "aisb-def-acceptance-criteria", "shape": "definition", "term": { "modality": "text", "value": "Acceptance criteria" }, "definition": { "modality": "text", "value": "Specific conditions an output must satisfy before it is considered usable." }, "tags": [ "prompting", "quality" ], "curatedDistractors": [ "Business context", "Source material", "First draft" ], "uid": "1bgo4dezqgnps" }, { "id": "aisb-mcq-better-prompt", "shape": "mcq", "prompt": { "modality": "text", "value": "Which request is most likely to produce a usable customer email?" }, "options": [ { "modality": "text", "value": "Write an email", "short": "Write an email" }, { "modality": "text", "value": "Write something friendly about our policy", "short": "Write friendly note re policy" }, { "modality": "text", "value": "Draft a 120-word reply to a delayed-order complaint using the policy below; acknowledge the delay, avoid promising a date, and offer two next steps", "short": "Draft 120-word reply, 2 steps" }, { "modality": "text", "value": "Act as the world's best marketer", "short": "Act as world's best marketer" } ], "correctIndex": 2, "explanation": "The strongest request supplies task, context, constraints, and a concrete output target.", "tags": [ "prompting", "context" ], "uid": "1nluks51ipq9f9" }, { "id": "aisb-proc-five-part-brief", "shape": "procedure", "goal": "Write a reusable AI brief", "steps": [ "State the task", "Describe the audience and business context", "Attach or paste trusted source material", "Set constraints and approval boundaries", "Define the output format and acceptance criteria" ], "notes": "Store successful briefs so the team does not reinvent them every time.", "tags": [ "prompting", "brief" ], "uid": "1pu648x19azexx" }, { "id": "aisb-fact-conversation", "shape": "fact", "title": "Use a Working Conversation", "body": "Strong results usually emerge through questions, critique, and revision. Treat the first answer as material to improve, not a final deliverable.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of two small-business colleagues reviewing and revising a draft together at a laptop, constructive back-and-forth conversation, natural office, no readable text, no logos.", "imageSearchTerm": "small business colleagues revising draft together", "alt": "Two colleagues discussing revisions to a draft", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Image_taken_from_page_101_of_%27(Pre-Adamite_Man;_or_the_story_of_our_old_planet,_etc._(By_Mrs._G._J._C._D.))%27_(11052735695).jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-conversation.webp" }, "tags": [ "prompting", "iteration" ], "uid": "t4ij1v1uruz6p" }, { "id": "aisb-fact-ask-questions", "shape": "fact", "title": "Let the AI Ask Questions", "body": "When important context is missing, instruct the AI to ask a few questions before drafting. This is especially useful for proposals, plans, and customer communications.", "tags": [ "prompting", "context" ], "illustration": { "imageSearchTerm": "conversation clarifying question desk", "imagePrompt": "Two people at a desk in conversation, one holding up a hand mid-question before the other begins writing on a notepad.", "alt": "Let the AI Ask Questions", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Boulder_Worldcup_Vienna_29-05-2010a_semifinals092_Akiyo_Noguchi.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-ask-questions.webp" }, "uid": "1o63k27ppujyj" }, { "id": "aisb-fact-save-pattern", "shape": "fact", "title": "Save the Pattern, Not Just the Output", "body": "When a workflow succeeds, save the instructions, source documents, examples, and review checklist. That package is more reusable than one finished draft.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead photograph of an organized reusable workflow kit with a binder, source documents, blank checklist, and laptop arranged together, realistic small-business desk, no readable text, no logos.", "imageSearchTerm": "organized business workflow binder checklist desk", "alt": "An organized reusable workflow kit on a desk", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:French_Creek_CCC.JPG", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-save-pattern.webp" }, "tags": [ "prompting", "reuse" ], "uid": "tqupxv1agr8zl" }, { "id": "aisb-pair-draft", "shape": "pair", "sideA": { "modality": "text", "value": "Draft pass" }, "sideB": { "modality": "text", "value": "Create a workable first version" }, "tags": [ "prompting", "iteration" ], "uid": "oh9sai1ijpo8k" }, { "id": "aisb-pair-critique", "shape": "pair", "sideA": { "modality": "text", "value": "Critique pass" }, "sideB": { "modality": "text", "value": "Identify weaknesses against explicit criteria" }, "tags": [ "prompting", "iteration" ], "uid": "zg1hkohgvujc" }, { "id": "aisb-def-prompt-library", "shape": "definition", "term": { "modality": "text", "value": "Prompt library" }, "definition": { "modality": "text", "value": "A maintained collection of tested instructions, examples, and checklists for recurring work." }, "tags": [ "prompting", "reuse" ], "curatedDistractors": [ "Brand guide", "Knowledge base", "Audit log" ], "uid": "16341xpqkb26n" }, { "id": "aisb-mcq-weak-output", "shape": "mcq", "prompt": { "modality": "text", "value": "An AI draft is generic and misses your differentiator. What is the best next move?" }, "options": [ { "modality": "text", "value": "Switch tools immediately", "short": "Switch tools immediately" }, { "modality": "text", "value": "Publish it and improve next month", "short": "Publish, improve next month" }, { "modality": "text", "value": "Add specific business context and a good example, then revise", "short": "Add context + example, revise" }, { "modality": "text", "value": "Ask for more enthusiasm", "short": "Ask for more enthusiasm" } ], "correctIndex": 2, "explanation": "Generic output usually signals missing context or examples, not necessarily the wrong tool.", "tags": [ "prompting", "iteration" ], "uid": "n0a6be1kmoarm" }, { "id": "aisb-proc-iteration-loop", "shape": "procedure", "goal": "Improve an AI draft", "steps": [ "Compare the draft with acceptance criteria", "Point out the most important gap", "Provide missing context or an example", "Request a focused revision", "Perform the final human edit and approval" ], "notes": "Change one important variable at a time so you learn what improves the result.", "tags": [ "prompting", "iteration" ], "uid": "956ufs1cz11tq" }, { "id": "aisb-fact-search-vs-memory", "shape": "fact", "title": "Current Questions Need Current Sources", "body": "For prices, laws, competitors, products, and market conditions, use an AI mode that searches the web and provides citations. Model memory alone may be stale.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business owner checking current market information on a laptop while comparing dated source documents, careful current research, no readable text, no logos.", "imageSearchTerm": "business owner researching current market information", "alt": "A business owner checking current market sources", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Business_registration_%E2%80%9EIngenieur_B%C3%BCro_Atlas_Kiel%22.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-search-vs-memory.webp" }, "tags": [ "research", "sources" ], "uid": "ks47zp1cpbsqt" }, { "id": "aisb-fact-primary-sources", "shape": "fact", "title": "Prefer Primary Sources", "body": "For material decisions, trace important claims to government data, original research, contracts, vendor documentation, or direct customer evidence.", "illustration": { "kind": "photo", "imagePrompt": "A business owner cross-referencing a printed contract against an official government document and vendor paperwork spread across a desk, no legible text.", "imageSearchTerm": "official documents research desk", "alt": "A business owner comparing original source documents", "depictable": true, "credit": "Unsplash · Anastassia Anufrieva · Unsplash License", "creditUrl": "https://unsplash.com/photos/a-woman-sitting-at-a-table-reading-a-paper-ecHGTPfjNfA", "subject": "Over-the-shoulder view of a woman in a dark jacket at a desk, holding one printed page up while reading another page in an open folder; more papers and blurred people across the table; no legible text", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-primary-sources.webp" }, "tags": [ "research", "sources" ], "uid": "1q8a0m7jwzm2j" }, { "id": "aisb-fact-survey-definitions", "shape": "fact", "title": "Definitions Change the Number", "body": "AI adoption estimates can differ because surveys ask different populations and define 'using AI' differently. The U.S. Census measured about 17–20% of businesses using AI in late 2025 through spring 2026, while a U.S. Chamber survey reported 58% of small businesses using generative AI in 2025.", "tags": [ "research", "statistics" ], "illustration": { "imageSearchTerm": "two different bar charts", "imagePrompt": "Two different bar charts side by side showing very different bar heights for what claims to be the same measurement, no legible text.", "alt": "Definitions Change the Number", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:One-two-two,_122_rue_de_Provence,_Paris_8e_1.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-survey-definitions.webp" }, "uid": "daz60dxwjf3t" }, { "id": "aisb-pair-primary-source", "shape": "pair", "sideA": { "modality": "text", "value": "Primary source" }, "sideB": { "modality": "text", "value": "Original data, law, policy, documentation, or firsthand evidence" }, "tags": [ "research", "sources" ], "uid": "37v1lt1qn8frf" }, { "id": "aisb-pair-secondary-source", "shape": "pair", "sideA": { "modality": "text", "value": "Secondary source" }, "sideB": { "modality": "text", "value": "A summary or interpretation of someone else's evidence" }, "tags": [ "research", "sources" ], "uid": "1snwc9m1owcwu" }, { "id": "aisb-def-triangulation", "shape": "definition", "term": { "modality": "text", "value": "Triangulation" }, "definition": { "modality": "text", "value": "Checking a material claim against multiple independent sources or methods before relying on it." }, "tags": [ "research", "verification" ], "curatedDistractors": [ "Summarization", "Brainstorming", "Forecasting" ], "uid": "eu7bc1yjb85o" }, { "id": "aisb-mcq-market-claim", "shape": "mcq", "prompt": { "modality": "text", "value": "An AI says local demand rose 40% this year. What evidence would be strongest?" }, "options": [ { "modality": "text", "value": "A vendor blog with no methodology", "short": "Vendor blog, no methodology" }, { "modality": "text", "value": "Current local sales, search, survey, or government data with a defined comparison period", "short": "Local sales/survey/gov data" }, { "modality": "text", "value": "A confident explanation of why demand rose", "short": "Confident why-it-rose story" }, { "modality": "text", "value": "A second AI repeating the claim", "short": "A second AI repeating it" } ], "correctIndex": 1, "explanation": "Decision-quality evidence needs a source, definition, location, and comparison period.", "tags": [ "research", "verification" ], "uid": "aho8wx1rszmzb" }, { "id": "aisb-proc-research-brief", "shape": "procedure", "goal": "Create a decision-ready research brief", "steps": [ "Write the decision the research must support", "List the claims that would change the decision", "Gather current sources and record their dates", "Separate evidence, assumptions, and estimates", "Verify the highest-impact claims independently", "Summarize implications and remaining uncertainty" ], "notes": "A research brief should make uncertainty visible rather than hiding it behind polished prose.", "tags": [ "research", "verification" ], "uid": "118kwrlhh2f5p" }, { "id": "aisb-fact-clean-data", "shape": "fact", "title": "Inspect Before You Analyze", "body": "Before asking for conclusions, identify missing values, duplicate rows, inconsistent dates, unclear units, and columns whose meaning is uncertain.", "illustration": { "kind": "photo", "imagePrompt": "A business analyst closely inspecting a spreadsheet on a laptop screen, highlighting a few irregular cells among the data, no legible text.", "imageSearchTerm": "analyst checking spreadsheet data", "alt": "An analyst inspecting spreadsheet data for problems", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:2012_CFA_Level_3_curriculum.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-clean-data.webp" }, "tags": [ "analysis", "data-quality" ], "uid": "1yjuxti1ygmx7e" }, { "id": "aisb-fact-observation-inference", "shape": "fact", "title": "Separate Observation from Inference", "body": "'Returns increased in June' is an observation. 'The new supplier caused the increase' is an inference that needs additional evidence.", "tags": [ "analysis", "reasoning" ], "illustration": { "imageSearchTerm": "returns boxes rising chart", "imagePrompt": "A small stack of returned packages in a store back room, with a rising bar chart on a nearby laptop screen showing the increase, no legible text.", "alt": "Separate Observation from Inference", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Tax_return.png", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-observation-inference.webp" }, "uid": "j70am18gobza" }, { "id": "aisb-fact-scenarios", "shape": "fact", "title": "Use Scenarios, Not One Forecast", "body": "A base case, upside case, and downside case reveal which assumptions drive the decision and where a contingency plan is needed.", "illustration": { "kind": "photo", "imagePrompt": "A business owner comparing three different printed forecast charts side by side on a table, representing low, middle, and high paths, no legible text.", "imageSearchTerm": "business forecast charts comparison", "alt": "A business owner comparing three planning scenarios", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Agreeing_on_a_business_plan_at_home.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-scenarios.webp" }, "tags": [ "analysis", "forecasting" ], "uid": "ub0yrh74ovar" }, { "id": "aisb-pair-observation", "shape": "pair", "sideA": { "modality": "text", "value": "Observation" }, "sideB": { "modality": "text", "value": "A pattern directly supported by the available data" }, "tags": [ "analysis", "reasoning" ], "uid": "66jd7e1g3g4qs" }, { "id": "aisb-pair-inference", "shape": "pair", "sideA": { "modality": "text", "value": "Inference" }, "sideB": { "modality": "text", "value": "An interpretation or explanation drawn from observations" }, "tags": [ "analysis", "reasoning" ], "uid": "7i8v2d1bc09x5" }, { "id": "aisb-def-sensitivity-analysis", "shape": "definition", "term": { "modality": "text", "value": "Sensitivity analysis" }, "definition": { "modality": "text", "value": "Testing how a result changes when an important assumption or input changes." }, "tags": [ "analysis", "forecasting" ], "curatedDistractors": [ "Trend analysis", "Data cleaning", "Reconciliation" ], "uid": "179gabs1qukqka" }, { "id": "aisb-mcq-data-first", "shape": "mcq", "prompt": { "modality": "text", "value": "What should happen before AI recommends actions from an uploaded sales spreadsheet?" }, "options": [ { "modality": "text", "value": "Inspect data quality and clarify column meanings", "short": "Check data quality + columns" }, { "modality": "text", "value": "Remove the lowest month", "short": "Remove the lowest month" }, { "modality": "text", "value": "Ask for the most surprising conclusion", "short": "Ask for most surprising result" }, { "modality": "text", "value": "Convert every value to a percentage", "short": "Convert all values to percents" } ], "correctIndex": 0, "explanation": "Bad or misunderstood input can produce convincing but useless analysis.", "tags": [ "analysis", "data-quality" ], "uid": "6yytjp1jswe0v" }, { "id": "aisb-proc-decision-memo", "shape": "procedure", "goal": "Build a one-page decision memo", "steps": [ "State the decision and deadline", "Summarize verified observations", "List assumptions and uncertainties", "Compare two or three realistic options", "Recommend an action with reasons", "Define the next measurement or checkpoint" ], "notes": "Keep the evidence visible so a reviewer can challenge the recommendation.", "tags": [ "analysis", "decision-making" ], "uid": "14omc9p1vj0t5r" }, { "id": "aisb-fact-brand-context", "shape": "fact", "title": "Give the Brand a Memory", "body": "A compact brand brief should define audience, promise, tone, proof points, prohibited claims, preferred terms, and examples of good work.", "illustration": { "kind": "photo", "imagePrompt": "Tasteful editorial photograph of a small-business team reviewing a cohesive brand board with color swatches, packaging samples, and campaign drafts, collaborative and practical, warm studio daylight, no readable text, no visible logos, no glossy stock-photo posing.", "imageSearchTerm": "small business team brand planning", "alt": "A small-business team reviewing a brand board", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Image_taken_from_page_124_of_%27La_Terra,_trattato_popolare_di_geografia_universale_per_G._Marinelli_ed_altri_scienziati_italiani,_etc._(With_illustrations_and_maps.)%27_(11157563895).jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-brand-context.webp" }, "tags": [ "marketing", "brand" ], "uid": "8fs8581o8mx3c" }, { "id": "aisb-fact-content-atom", "shape": "fact", "title": "Create Once, Adapt Carefully", "body": "One useful customer story or guide can become an email, several social posts, a sales handout, and short video scripts when each version is adapted to its channel.", "tags": [ "marketing", "content" ], "illustration": { "imageSearchTerm": "content repurposed multiple formats", "imagePrompt": "A single printed customer story on a desk surrounded by a phone showing a social post, a printed flyer, and a small storyboard, all adapted from it, no legible text.", "alt": "Create Once, Adapt Carefully", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Eli_Content_met_gedicht_voor_raam_eigen_atelier,_Weesperzijde_41,_A%27dam_1988.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-content-atom.webp" }, "uid": "1ufapdu1l308o0" }, { "id": "aisb-fact-copyright", "shape": "fact", "title": "Human Authorship Still Matters", "body": "The U.S. Copyright Office says AI-assisted work can qualify for protection when humans contribute sufficient expressive authorship, but prompts alone generally do not make a purely AI-generated output copyrightable.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of a human illustrator actively drawing an original composition by hand at a studio desk while a laptop remains a secondary reference tool, clear human creative authorship, no readable text, no logos.", "imageSearchTerm": "human illustrator drawing original artwork studio laptop", "alt": "A human illustrator creating original artwork by hand", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:01764_4_layers_Human_Tower_2011_Rio_de_Janeiro.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-copyright.webp" }, "tags": [ "marketing", "copyright" ], "curatedDistractors": [ "Federal Trade Commission", "Small Business Administration", "National Institute of Standards and Technology" ], "uid": "ypobbt1bh720t" }, { "id": "aisb-pair-brand-voice", "shape": "pair", "sideA": { "modality": "text", "value": "Brand voice" }, "sideB": { "modality": "text", "value": "The consistent personality and language choices used in communication" }, "tags": [ "marketing", "brand" ], "uid": "1byx89hs05ncn" }, { "id": "aisb-pair-proof-point", "shape": "pair", "sideA": { "modality": "text", "value": "Proof point" }, "sideB": { "modality": "text", "value": "Specific evidence supporting a marketing claim" }, "tags": [ "marketing", "claims" ], "uid": "1d7d9yhlazedt" }, { "id": "aisb-def-content-atom", "shape": "definition", "term": { "modality": "text", "value": "Content atom" }, "definition": { "modality": "text", "value": "A substantial source asset that can be adapted into multiple smaller channel-specific pieces." }, "tags": [ "marketing", "content" ], "curatedDistractors": [ "Brand brief", "Proof point", "Editorial calendar" ], "uid": "1t9po4q1a5hogw" }, { "id": "aisb-mcq-claim-review", "shape": "mcq", "prompt": { "modality": "text", "value": "AI drafts an advertisement claiming your service is 'guaranteed to double sales.' What should you do?" }, "options": [ { "modality": "text", "value": "Remove or substantiate the claim and review it for legal and brand risk", "short": "Remove/prove claim, check risk" }, { "modality": "text", "value": "Publish it if it sounds persuasive", "short": "Publish if it's persuasive" }, { "modality": "text", "value": "Replace 'guaranteed' with 'almost certain'", "short": "Swap 'guaranteed' for 'almost'" }, { "modality": "text", "value": "Ask AI to add a testimonial", "short": "Ask AI to add a testimonial" } ], "correctIndex": 0, "explanation": "Marketing claims need truthful support. AI does not make an unsupported promise safe.", "tags": [ "marketing", "claims" ], "uid": "yklsyq1awjyp2" }, { "id": "aisb-proc-content-workflow", "shape": "procedure", "goal": "Turn one source into a small campaign", "steps": [ "Choose an original source with a clear customer lesson", "Extract the main message and supporting proof", "Draft channel-specific versions", "Check facts, claims, links, and brand voice", "Have an owner approve external content", "Measure response and record what worked" ], "notes": "Repurposing should preserve the core truth while adapting format and emphasis.", "tags": [ "marketing", "content" ], "uid": "16jhl6n1rsbqkt" }, { "id": "aisb-fact-approved-knowledge", "shape": "fact", "title": "Ground Customer Answers", "body": "A customer assistant should answer from approved policies, product information, hours, pricing rules, and escalation instructions—not from general model knowledge.", "illustration": { "kind": "photo", "imagePrompt": "Authentic editorial photograph of a customer-service employee at a small local business consulting an approved reference binder while replying on a computer, attentive and calm, natural workplace lighting, no readable customer data, no logos, no chatbot graphics.", "imageSearchTerm": "small business customer service employee computer", "alt": "A customer-service employee consulting approved information", "credit": "Pexels · Kampus Production · Pexels License", "creditUrl": "https://www.pexels.com/photo/woman-in-gray-sweater-working-on-a-desktop-computers-8204323/", "subject": "A woman in a headset seated at a desk, hand to her earpiece, looking at a monitor showing a dashboard of charts and tables, keyboard and papers in front of her, industrial office blurred behind.", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-approved-knowledge.webp" }, "tags": [ "customer-service", "knowledge" ], "uid": "6citmzca76a9" }, { "id": "aisb-fact-escalate", "shape": "fact", "title": "Design the Handoff", "body": "Automation is safer when it knows when to stop. Complaints, refunds above a limit, threats, legal questions, and unusual exceptions should route to a person.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of a customer-service employee handing a complex case folder to an experienced supervisor, clear human escalation in a small office, no readable text, no logos.", "imageSearchTerm": "customer service employee escalating case supervisor", "alt": "A customer-service employee handing a complex case to a supervisor", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Beispiel_Customer_Journey_Map.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-escalate.webp" }, "tags": [ "customer-service", "escalation" ], "uid": "5miuhnyytcqj" }, { "id": "aisb-fact-personalization", "shape": "fact", "title": "Personalize with Restraint", "body": "Useful personalization reflects relevant customer context. Creepy personalization reveals data the customer did not expect you to use.", "tags": [ "sales", "privacy" ], "illustration": { "imageSearchTerm": "handwritten note package customer", "imagePrompt": "A friendly handwritten note with a customer's first name tucked into a shipped package, beside a separate stack of detailed customer files kept firmly closed.", "alt": "Personalize with Restraint", "depictable": false, "credit": "Pexels · RDNE Stock project · Pexels License", "creditUrl": "https://www.pexels.com/photo/healthy-marketing-people-connection-7310245/", "subject": "two hands with bracelets placing a printed 'Thank you so much' card onto folded pale garments inside an open cardboard shipping box", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-personalization.webp" }, "uid": "17btirv1bkqhkd" }, { "id": "aisb-pair-qualification", "shape": "pair", "sideA": { "modality": "text", "value": "Lead qualification" }, "sideB": { "modality": "text", "value": "Checking whether a prospect matches defined needs and buying conditions" }, "tags": [ "sales", "leads" ], "uid": "1v85ydna8ywar" }, { "id": "aisb-pair-escalation", "shape": "pair", "sideA": { "modality": "text", "value": "Escalation" }, "sideB": { "modality": "text", "value": "Routing a case to a person with the authority or expertise to resolve it" }, "tags": [ "customer-service", "escalation" ], "uid": "15uo7ili82bhj" }, { "id": "aisb-def-escalation-rule", "shape": "definition", "term": { "modality": "text", "value": "Escalation rule" }, "definition": { "modality": "text", "value": "A defined condition that stops automation and transfers a case to human review." }, "tags": [ "customer-service", "escalation" ], "curatedDistractors": [ "Approval gate", "Exception path", "Kill switch" ], "uid": "1iyif0h12gkoob" }, { "id": "aisb-mcq-chatbot-unknown", "shape": "mcq", "prompt": { "modality": "text", "value": "A chatbot cannot find an approved answer to a customer's warranty question. What should it do?" }, "options": [ { "modality": "text", "value": "End the conversation without explanation", "short": "End chat with no explanation" }, { "modality": "text", "value": "Say it is unsure and escalate to a person", "short": "Say unsure, escalate to person" }, { "modality": "text", "value": "Give a generic industry answer", "short": "Give a generic industry answer" }, { "modality": "text", "value": "Invent the most likely policy", "short": "Invent the most likely policy" } ], "correctIndex": 1, "explanation": "An explicit uncertainty response and handoff protects trust better than a plausible invention.", "tags": [ "customer-service", "escalation" ], "uid": "ipbddy1ar3ark" }, { "id": "aisb-proc-faq-assistant", "shape": "procedure", "goal": "Prepare a customer-answer workflow", "steps": [ "Collect the most common questions", "Write and approve canonical answers", "Identify information the assistant may access", "Define prohibited promises and escalation triggers", "Test normal, ambiguous, and adversarial questions", "Review logs and update the knowledge regularly" ], "notes": "Start in draft-only mode before allowing automatic replies.", "tags": [ "customer-service", "automation" ], "uid": "1k6oidc36h52a" }, { "id": "aisb-fact-capture-process", "shape": "fact", "title": "Capture the Real Process", "body": "The fastest way to draft a procedure is often to narrate the task while doing it, then have AI organize the transcript into steps, decisions, and exceptions.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of an experienced small-business worker demonstrating a hands-on process while a colleague documents each step on a tablet, realistic workshop or service environment, practical and human, no visible logos, no readable text, no staged corporate look.", "imageSearchTerm": "small business worker documenting process", "alt": "A worker demonstrating a process while a colleague documents it", "credit": "Unsplash · 毛 祥 · Unsplash License", "creditUrl": "https://unsplash.com/photos/man-in-green-zip-up-jacket-using-smartphone-yRrTJZSNmYw", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-capture-process.webp" }, "tags": [ "operations", "sop" ], "uid": "1m204pj1uj1l1b" }, { "id": "aisb-fact-exceptions", "shape": "fact", "title": "Exceptions Are Part of the Procedure", "body": "A clean happy-path checklist is incomplete. Good operating documentation explains what to do when information is missing, a customer objects, or a system fails.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business team reviewing a process map with several branching exception paths marked by colored sticky notes, realistic planning session, no readable text, no logos.", "imageSearchTerm": "business team process exceptions sticky notes", "alt": "A team reviewing branching exceptions in a business process", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Drag-along_right_(DAR).jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-exceptions.webp" }, "tags": [ "operations", "exceptions" ], "uid": "vq7z0d1u3z663" }, { "id": "aisb-fact-meeting-action", "shape": "fact", "title": "Meetings Need Owners and Deadlines", "body": "AI summaries become operationally useful when decisions, action items, owners, due dates, and unresolved questions are separated clearly.", "tags": [ "operations", "meetings" ], "illustration": { "imageSearchTerm": "whiteboard meeting sticky notes", "imagePrompt": "A conference room whiteboard after a meeting, showing separate columns of sticky notes for decisions, owners, and deadlines, no legible text.", "alt": "Meetings Need Owners and Deadlines", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Alignment_Shuffling_for_Identity.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-meeting-action.webp" }, "uid": "dkpe0c1wkjpk6" }, { "id": "aisb-pair-happy-path", "shape": "pair", "sideA": { "modality": "text", "value": "Happy path" }, "sideB": { "modality": "text", "value": "The normal sequence when nothing unusual happens" }, "tags": [ "operations", "sop" ], "uid": "ejt5w1xd7467" }, { "id": "aisb-pair-exception-path", "shape": "pair", "sideA": { "modality": "text", "value": "Exception path" }, "sideB": { "modality": "text", "value": "Instructions for unusual conditions or failures" }, "tags": [ "operations", "sop" ], "uid": "6hooftvan7e1" }, { "id": "aisb-def-sop", "shape": "definition", "term": { "modality": "text", "value": "Standard operating procedure" }, "definition": { "modality": "text", "value": "A maintained description of how to perform a recurring task, including steps, responsibilities, controls, and exceptions." }, "tags": [ "operations", "sop" ], "curatedDistractors": [ "Policy", "Decision memo", "Project brief" ], "uid": "1gasu1x1bpb2i5" }, { "id": "aisb-mcq-sop-source", "shape": "mcq", "prompt": { "modality": "text", "value": "What is the strongest source for a new procedure?" }, "options": [ { "modality": "text", "value": "A generic AI answer about the industry", "short": "A generic AI industry answer" }, { "modality": "text", "value": "The actual process demonstrated by an experienced worker", "short": "Process shown by expert worker" }, { "modality": "text", "value": "A competitor's marketing page", "short": "A competitor's marketing page" }, { "modality": "text", "value": "An old checklist no one follows", "short": "Old checklist no one follows" } ], "correctIndex": 1, "explanation": "AI can organize knowledge, but the source should be the real process and people who understand it.", "tags": [ "operations", "sop" ], "uid": "1q3jzx11kdyk6j" }, { "id": "aisb-proc-build-sop", "shape": "procedure", "goal": "Create and test an AI-assisted procedure", "steps": [ "Observe or record the task being performed", "Draft the steps, decisions, inputs, and outputs", "Add safety checks and exception paths", "Have an experienced person review it", "Ask a different person to follow it", "Revise and assign an owner and review date" ], "notes": "A procedure is not finished until someone else can use it successfully.", "tags": [ "operations", "sop" ], "uid": "2npra21sj6n92" }, { "id": "aisb-fact-trigger-action", "shape": "fact", "title": "Automation Starts with a Trigger", "body": "A basic automation follows a rule: when a defined event occurs, perform a defined action. AI can classify or draft inside that flow, but the surrounding controls still matter.", "tags": [ "automation", "workflow" ], "illustration": { "imageSearchTerm": "automated switch triggering machine", "imagePrompt": "A mechanical switch or sensor triggering a small conveyor belt into motion, representing an automated rule responding to an event.", "alt": "Automation Starts with a Trigger", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Automated_Planet_Finder_Dome.JPG", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-trigger-action.webp" }, "uid": "2933uqds4kjw" }, { "id": "aisb-fact-agent-permission", "shape": "fact", "title": "Agents Need Boundaries", "body": "An agent may choose steps and use tools toward a goal. Limit what it can read, change, send, spend, or delete, and require approval for consequential actions.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business owner reviewing an automation approval screen before authorizing an action, thoughtful posture, ordinary modern office, emphasis on human control rather than futuristic AI, no readable interface text, no logos, no robots or holograms.", "imageSearchTerm": "business owner reviewing automation approval", "alt": "A business owner reviewing an automated action before approval", "credit": "Pexels · RDNE Stock project · Pexels License", "creditUrl": "https://www.pexels.com/photo/businessman-signing-documents-in-his-office-10376251/", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-agent-permission.webp" }, "tags": [ "automation", "agents" ], "uid": "hlrjct9yipb" }, { "id": "aisb-fact-logs", "shape": "fact", "title": "Keep an Audit Trail", "body": "Logs make it possible to see what ran, what information was used, what the AI proposed, what action occurred, and who approved it.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead photograph of a business owner reviewing an orderly chronological activity log beside approval records and a laptop, accountability and traceability, no readable text, no logos.", "imageSearchTerm": "business owner reviewing audit log records", "alt": "A business owner reviewing activity and approval records", "credit": "Unsplash · Vitaly Gariev · Unsplash License", "creditUrl": "https://unsplash.com/photos/a-man-sitting-at-a-desk-working-on-a-computer-Oexx7cEMKFA", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-logs.webp" }, "tags": [ "automation", "governance" ], "uid": "11p35x61ez2m7g" }, { "id": "aisb-pair-automation", "shape": "pair", "sideA": { "modality": "text", "value": "Rule-based automation" }, "sideB": { "modality": "text", "value": "A predefined trigger and action sequence" }, "tags": [ "automation", "workflow" ], "uid": "5v6d3ojr707q" }, { "id": "aisb-pair-agent", "shape": "pair", "sideA": { "modality": "text", "value": "AI agent" }, "sideB": { "modality": "text", "value": "A system that can select steps and use tools toward a goal within permissions" }, "tags": [ "automation", "agents" ], "uid": "k5uitk1d8mmfu" }, { "id": "aisb-def-approval-gate", "shape": "definition", "term": { "modality": "text", "value": "Approval gate" }, "definition": { "modality": "text", "value": "A required human confirmation before a workflow performs a consequential action." }, "tags": [ "automation", "governance" ], "curatedDistractors": [ "Audit log", "Escalation rule", "Kill switch" ], "uid": "5a9lp917vj6z9" }, { "id": "aisb-mcq-first-agent", "shape": "mcq", "prompt": { "modality": "text", "value": "Which action should normally require approval in a beginner business automation?" }, "options": [ { "modality": "text", "value": "Extracting dates from a document", "short": "Extracting dates from a doc" }, { "modality": "text", "value": "Labeling an email for review", "short": "Labeling an email for review" }, { "modality": "text", "value": "Sending a refund from the business account", "short": "Sending a refund from account" }, { "modality": "text", "value": "Drafting an internal summary", "short": "Drafting an internal summary" } ], "correctIndex": 2, "explanation": "Moving money is consequential and should stay behind a human approval gate.", "tags": [ "automation", "risk" ], "uid": "1f5zq2k1d9uvoy" }, { "id": "aisb-proc-automate-safely", "shape": "procedure", "goal": "Move a workflow toward automation", "steps": [ "Document the current workflow", "Remove unnecessary steps and clarify rules", "Run AI in draft-only mode", "Measure errors and identify exceptions", "Add permissions, approval gates, and logs", "Increase autonomy gradually and keep a kill switch" ], "notes": "Do not automate a process you cannot explain or evaluate.", "tags": [ "automation", "governance" ], "uid": "iu950rx7swyz" }, { "id": "aisb-fact-first-pass", "shape": "fact", "title": "First Pass, Not Final Authority", "body": "AI can summarize a contract, organize receipts, draft questions, or flag unusual terms. It should not be treated as the final authority on legal, accounting, lending, or tax decisions.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business owner using a laptop summary as a first pass while a professional reviews the original contract documents, clear human authority, no readable text, no logos.", "imageSearchTerm": "business owner professional reviewing contract documents", "alt": "A professional reviewing original documents with a business owner", "credit": "Pexels · Pavel Danilyuk · Pexels License", "creditUrl": "https://www.pexels.com/photo/a-businessman-reviewing-some-documents-7519014/", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-first-pass.webp" }, "tags": [ "high-stakes", "review" ], "uid": "1komqq1eeiwfw" }, { "id": "aisb-fact-reconcile", "shape": "fact", "title": "Reconcile Against the Source", "body": "Financial summaries should be tied back to statements, invoices, payroll records, or the accounting system. A plausible total is not a reconciled total.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead photograph of a small-business bookkeeper matching a summary against invoices, receipts, and account statements with a calculator, careful reconciliation, no readable financial data, no logos.", "imageSearchTerm": "small business bookkeeper reconciling invoices statements", "alt": "A bookkeeper matching a financial summary to source records", "credit": "Pexels · www.kaboompics.com · Pexels License", "creditUrl": "https://www.pexels.com/photo/a-woman-computing-bills-while-holding-a-pencil-7680742/", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-reconcile.webp" }, "tags": [ "finance", "verification" ], "uid": "183l58fsgc2xh" }, { "id": "aisb-fact-professional", "shape": "fact", "title": "Use Professionals Where Judgment Matters", "body": "An attorney, CPA, bookkeeper, insurance professional, or lender can evaluate facts and obligations that a general AI assistant may miss.", "tags": [ "high-stakes", "professional" ], "illustration": { "imageSearchTerm": "small business owner consulting attorney", "imagePrompt": "A small business owner sits across a desk from an attorney, reviewing paperwork together in an office.", "alt": "Use Professionals Where Judgment Matters", "depictable": true }, "uid": "drulkr12ei92h" }, { "id": "aisb-pair-first-pass", "shape": "pair", "sideA": { "modality": "text", "value": "AI first pass" }, "sideB": { "modality": "text", "value": "Summarize, classify, compare, or prepare questions" }, "tags": [ "high-stakes", "review" ], "uid": "1tc01oi1s4b7oc" }, { "id": "aisb-pair-final-decision", "shape": "pair", "sideA": { "modality": "text", "value": "Qualified human decision" }, "sideB": { "modality": "text", "value": "Accept responsibility for advice, filing, approval, or commitment" }, "tags": [ "high-stakes", "review" ], "uid": "wzu98y16vzroc" }, { "id": "aisb-def-material-term", "shape": "definition", "term": { "modality": "text", "value": "Material term" }, "definition": { "modality": "text", "value": "A contract provision important enough that it could affect cost, rights, obligations, risk, or the decision to sign." }, "tags": [ "high-stakes", "contracts" ], "curatedDistractors": [ "Formatting note", "Nonbinding example", "Document heading" ], "uid": "ynb7ppei6rc5" }, { "id": "aisb-mcq-lease", "shape": "mcq", "prompt": { "modality": "text", "value": "AI flags an unusual indemnity clause in a commercial lease. What is the best next step?" }, "options": [ { "modality": "text", "value": "Ignore it because AI sometimes makes mistakes", "short": "Ignore it; AI errs sometimes" }, { "modality": "text", "value": "Ask AI to guarantee the lease is safe", "short": "Ask AI to guarantee it's safe" }, { "modality": "text", "value": "Delete the clause yourself and sign", "short": "Delete the clause, then sign" }, { "modality": "text", "value": "Use the flag to prepare questions for a qualified attorney", "short": "Prep questions for an attorney" } ], "correctIndex": 3, "explanation": "AI can focus professional review, but the legal judgment and negotiation should come from a qualified person.", "tags": [ "high-stakes", "contracts" ], "uid": "i3alz312siw0f" }, { "id": "aisb-proc-high-stakes-review", "shape": "procedure", "goal": "Use AI for high-stakes preparation", "steps": [ "Use an approved tool and remove unnecessary sensitive data", "Ask for a structured summary with cited sections", "List assumptions, missing information, and questions", "Compare important figures or clauses with original records", "Send the source and questions to the responsible professional", "Record the final human decision" ], "notes": "Never let a summary replace the underlying document or record.", "tags": [ "high-stakes", "review" ], "uid": "vp81q1lywu9o" }, { "id": "aisb-fact-employment-law", "shape": "fact", "title": "Existing Employment Law Still Applies", "body": "Using an algorithm does not remove an employer's obligations. AI-assisted recruiting or assessment can still create unlawful discrimination.", "illustration": { "kind": "photo", "imagePrompt": "A hiring manager reviews a stack of printed resumes at a desk with a computer screen showing candidate profiles.", "imageSearchTerm": "hiring manager reviewing resumes", "alt": "A diverse hiring team reviewing candidates with a compliance checklist", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Old_Prison_-_panoramio_-_Halooch.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-employment-law.webp" }, "tags": [ "people", "hiring" ], "uid": "1c63vk57d2bll" }, { "id": "aisb-fact-accommodation", "shape": "fact", "title": "Plan for Accommodation", "body": "The EEOC advises employers to consider whether algorithmic tools screen out people with disabilities and to provide a process for reasonable accommodation.", "tags": [ "people", "accessibility" ], "illustration": { "imageSearchTerm": "wheelchair accessible office workstation", "imagePrompt": "An employee using a wheelchair works at an adjustable-height desk in a modern office.", "alt": "Plan for Accommodation", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Dyke_Access_Trail_-_geograph.org.uk_-_36803.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-accommodation.webp" }, "uid": "1kos6p81udvy3y" }, { "id": "aisb-fact-human-decision", "shape": "fact", "title": "Keep Meaningful Human Review", "body": "AI can organize applications against documented job requirements, but a responsible person should review context, challenge errors, and make employment decisions.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of a hiring manager thoughtfully comparing several anonymized candidate folders against a blank requirements checklist after a laptop-assisted sorting step, visible human judgment, all pages turned away or fully unreadable, no words, letters, numbers, personal information, or logos.", "imageSearchTerm": "hiring manager reviewing applications job requirements", "alt": "A hiring manager personally reviewing applications", "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Old_Prison_-_panoramio.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-human-decision.webp" }, "tags": [ "people", "hiring" ], "uid": "5o2wj51kb76h7" }, { "id": "aisb-pair-objective-criterion", "shape": "pair", "sideA": { "modality": "text", "value": "Objective criterion" }, "sideB": { "modality": "text", "value": "A job-related requirement defined before reviewing candidates" }, "tags": [ "people", "hiring" ], "uid": "3k6kfzgw2ak1" }, { "id": "aisb-pair-subjective-judgment", "shape": "pair", "sideA": { "modality": "text", "value": "Subjective judgment" }, "sideB": { "modality": "text", "value": "An interpretation that can be influenced by assumptions or bias" }, "tags": [ "people", "hiring" ], "uid": "rxk0llrk62j" }, { "id": "aisb-def-adverse-impact", "shape": "definition", "term": { "modality": "text", "value": "Adverse impact" }, "definition": { "modality": "text", "value": "A substantially different negative outcome for a protected group caused by a seemingly neutral employment practice." }, "tags": [ "people", "fairness" ], "curatedDistractors": [ "Reasonable accommodation", "Objective criterion", "Individual review" ], "uid": "1thmsllbajfud" }, { "id": "aisb-mcq-resume-screen", "shape": "mcq", "prompt": { "modality": "text", "value": "What is the safest beginner use of AI in recruiting?" }, "options": [ { "modality": "text", "value": "Rank candidates by names and photos", "short": "Rank by names and photos" }, { "modality": "text", "value": "Make the final hiring decision", "short": "Make the final hiring call" }, { "modality": "text", "value": "Draft a job description from documented role requirements for human review", "short": "Draft job desc, human review" }, { "modality": "text", "value": "Automatically reject applicants based on personality predictions", "short": "Auto-reject on personality" } ], "correctIndex": 2, "explanation": "Drafting from documented requirements is administrative and reviewable; automated judgments about people carry much greater risk.", "tags": [ "people", "hiring" ], "uid": "wk009gwpvm2g" }, { "id": "aisb-proc-hiring-aid", "shape": "procedure", "goal": "Use AI as a hiring aid", "steps": [ "Define job-related requirements and evaluation criteria", "Remove criteria unrelated to successful performance", "Use AI for drafting or organization, not final decisions", "Check outputs for unsupported assumptions and disparate treatment", "Provide accommodation and alternative-process information", "Document the human review and decision" ], "notes": "Consult current federal, state, and local requirements for your location and workforce.", "tags": [ "people", "hiring" ], "uid": "1x5fe1fjjhuf3" }, { "id": "aisb-fact-data-minimization", "shape": "fact", "title": "Share the Minimum", "body": "Do not paste an entire customer file when the task only needs a product category and date. Data minimization reduces exposure and often improves focus.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead photograph of a small-business owner practicing data minimization: a thick closed customer-record folder remains set aside while the owner transfers only two simple non-identifying details onto a small blank note beside a laptop, privacy-conscious realistic workspace, no screens visible, no readable personal data, no logos, no floating icons.", "imageSearchTerm": "business owner organizing customer records privacy", "alt": "A business owner setting aside a customer file and using only a small note", "credit": "AI-generated (gpt-image-1.5)", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-data-minimization.webp" }, "tags": [ "privacy", "data" ], "uid": "1th2vun1dy1ykt" }, { "id": "aisb-fact-plan-controls", "shape": "fact", "title": "Plan Labels Are Not Data Controls", "body": "Free, personal, business, and enterprise products can have different training, retention, sharing, and administration settings. Verify the current terms and controls for the exact product you use.", "illustration": { "kind": "photo", "imagePrompt": "Editorial photograph of a small-business owner comparing three colored, completely blank plan folders, each paired with a different number of small physical padlocks and administrator keycards to represent different privacy controls, realistic desk, no words, letters, numbers, screens, product names, or logos anywhere.", "imageSearchTerm": "business owner comparing software privacy plans", "alt": "A business owner comparing colored plan folders with different physical padlocks" }, "tags": [ "privacy", "vendors" ], "uid": "dyssap4iyhrt" }, { "id": "aisb-fact-business-data", "shape": "fact", "title": "Business Products May Offer Stronger Defaults", "body": "For example, OpenAI states that ChatGPT Business, Enterprise, and API data are not used to train models by default. Every vendor should be evaluated separately before sensitive use.", "tags": [ "privacy", "vendors" ], "curatedDistractors": [ "Claude Team", "Gemini for Workspace", "Microsoft 365 Copilot" ], "illustration": { "imageSearchTerm": "business data privacy security", "imagePrompt": "A glowing padlock shields a stream of abstract data icons flowing toward a cloud server, symbolizing protected business data.", "alt": "Business Products May Offer Stronger Defaults", "depictable": false }, "uid": "17zpikd1vjh30j" }, { "id": "aisb-pair-public-data", "shape": "pair", "sideA": { "modality": "text", "value": "Public data" }, "sideB": { "modality": "text", "value": "Information intentionally available to anyone" }, "tags": [ "privacy", "classification" ], "uid": "1tf74xp1auhmlf" }, { "id": "aisb-pair-confidential-data", "shape": "pair", "sideA": { "modality": "text", "value": "Confidential data" }, "sideB": { "modality": "text", "value": "Restricted business, customer, employee, or partner information" }, "tags": [ "privacy", "classification" ], "uid": "1pouuam1rt1ogq" }, { "id": "aisb-def-data-retention", "shape": "definition", "term": { "modality": "text", "value": "Data retention" }, "definition": { "modality": "text", "value": "How long a service or business keeps information before deletion or archival." }, "tags": [ "privacy", "data" ], "curatedDistractors": [ "Data minimization", "Access control", "Encryption" ], "uid": "1l29ebdqqwyjt" }, { "id": "aisb-mcq-customer-list", "shape": "mcq", "prompt": { "modality": "text", "value": "You want AI to segment a customer list. What should happen first?" }, "options": [ { "modality": "text", "value": "Verify authorization, tool controls, needed fields, and retention before sharing data", "short": "Check consent, controls, fields" }, { "modality": "text", "value": "Replace customer names but keep every other identifier", "short": "Mask names, keep other IDs" }, { "modality": "text", "value": "Ask the AI provider whether breaches are impossible", "short": "Ask if breaches are impossible" }, { "modality": "text", "value": "Upload the full list to a free tool", "short": "Upload full list to free tool" } ], "correctIndex": 0, "explanation": "Sensitive-data work starts with purpose, authorization, minimization, vendor controls, and retention.", "tags": [ "privacy", "data" ], "uid": "1mev7f91wc4iv1" }, { "id": "aisb-proc-data-check", "shape": "procedure", "goal": "Check data before using an AI tool", "steps": [ "Name the business purpose", "Classify the information involved", "Remove fields that are not necessary", "Verify the tool's current training, retention, access, and sharing controls", "Confirm customer, employee, contractual, and legal permissions", "Choose a safer method or stop if controls are insufficient" ], "notes": "When in doubt, use synthetic or de-identified examples for experimentation.", "tags": [ "privacy", "governance" ], "uid": "54u4sx1vpsg6v" }, { "id": "aisb-fact-risk-tier", "shape": "fact", "title": "Control by Consequence", "body": "The more an AI output can affect money, rights, safety, reputation, or sensitive data, the stronger the testing, approval, documentation, and monitoring should be.", "illustration": { "kind": "photo", "imagePrompt": "Editorial overhead still life of three completely unlabeled work trays colored green, yellow, and red. The green tray holds blank marketing cards, the yellow tray holds a closed customer folder with a small padlock, and the red tray holds a calculator and payment card beneath a person's hand applying a blank approval stamp, realistic desk, no words, letters, numbers, screens, symbols, or logos.", "imageSearchTerm": "business risk levels human approval", "alt": "Green, yellow, and red work trays with increasing safeguards and human approval", "credit": "AI-generated (gpt-image-1.5)", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-risk-tier.webp" }, "tags": [ "governance", "risk" ], "uid": "1yotgnd1nke7bt" }, { "id": "aisb-fact-nist-functions", "shape": "fact", "title": "Govern, Map, Measure, Manage", "body": "NIST organizes AI risk work into four functions: establish governance, understand the context, assess risks and performance, and prioritize responses.", "tags": [ "governance", "risk" ], "illustration": { "imageSearchTerm": "AI governance risk framework", "imagePrompt": "Four interlocking gears are arranged in a circular flow, each glowing softly to represent a different stage of a risk management cycle.", "alt": "Govern, Map, Measure, Manage", "depictable": false, "credit": "Unsplash · Sasun Bughdaryan · Unsplash License", "creditUrl": "https://unsplash.com/photos/blue-blocks-spelling-risk-next-to-a-magnifying-glass-q5kwAqdyRe8", "subject": "four blue cubes stacked vertically spelling RISK beside a handheld magnifying glass on a pale blue backdrop", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-nist-functions.webp" }, "uid": "1xh0jotgao41r" }, { "id": "aisb-fact-incidents", "shape": "fact", "title": "Plan for Failure", "body": "An AI incident plan should say how to stop the workflow, preserve logs, correct affected records or messages, notify responsible people, and prevent recurrence.", "tags": [ "governance", "incidents" ], "illustration": { "imageSearchTerm": "IT staff responding server alert", "imagePrompt": "An IT professional in a server room responds urgently to a flashing red alert on a monitor.", "alt": "Plan for Failure", "depictable": true, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Logo_M.P._%26_Records.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-incidents.webp" }, "uid": "1ytb0dy146cbpe" }, { "id": "aisb-pair-reversible", "shape": "pair", "sideA": { "modality": "text", "value": "Reversible action" }, "sideB": { "modality": "text", "value": "An action that can be corrected with limited harm or cost" }, "tags": [ "governance", "risk" ], "uid": "1nujfn119469cf" }, { "id": "aisb-pair-irreversible", "shape": "pair", "sideA": { "modality": "text", "value": "Consequential action" }, "sideB": { "modality": "text", "value": "An action that can materially affect money, rights, safety, or trust" }, "tags": [ "governance", "risk" ], "uid": "785xdiw5vw4y" }, { "id": "aisb-def-human-in-loop", "shape": "definition", "term": { "modality": "text", "value": "Human in the loop" }, "definition": { "modality": "text", "value": "A workflow design in which a person reviews, approves, corrects, or can stop AI-supported work at a meaningful point." }, "tags": [ "governance", "oversight" ], "curatedDistractors": [ "Audit log", "Kill switch", "Automated trigger" ], "uid": "1bl02e21dsh626" }, { "id": "aisb-mcq-review-level", "shape": "mcq", "prompt": { "modality": "text", "value": "Which workflow needs the strongest review and controls?" }, "options": [ { "modality": "text", "value": "Drafting three subject-line options", "short": "Drafting 3 subject-line options" }, { "modality": "text", "value": "Generating internal brainstorming ideas", "short": "Generating brainstorm ideas" }, { "modality": "text", "value": "Reformatting a meeting agenda", "short": "Reformatting a meeting agenda" }, { "modality": "text", "value": "Automatically changing employee pay based on an AI score", "short": "Auto-changing pay by AI score" } ], "correctIndex": 3, "explanation": "Pay decisions affect rights and livelihoods, so they require strong controls and accountable human judgment.", "tags": [ "governance", "risk" ], "uid": "1stv2wh1q7eofh" }, { "id": "aisb-proc-preflight", "shape": "procedure", "goal": "Run an AI workflow preflight", "steps": [ "Define the intended use and owner", "Identify affected people, data, and decisions", "Rate the consequence of errors or misuse", "Set tests, approval gates, permissions, and logs", "Pilot with limited scope", "Monitor outcomes and document incidents and changes" ], "notes": "Reassess the workflow when the model, vendor, data, or business process changes.", "tags": [ "governance", "quality" ], "uid": "37jxo21lmtja" }, { "id": "aisb-fact-one-pilot", "shape": "fact", "title": "One Pilot Beats Ten Trials", "body": "A focused 30-day pilot creates evidence. Tool hopping creates opinions without a stable process or baseline.", "illustration": { "kind": "photo", "imagePrompt": "Natural editorial photograph of a small-business owner reviewing a simple thirty-day pilot plan with one focused workflow, calendar and measurement notes visible only as abstract marks, optimistic but disciplined mood, warm daylight, no readable text, no logos, no futuristic imagery.", "imageSearchTerm": "small business owner pilot project planning", "alt": "A small-business owner reviewing a focused pilot plan", "credit": "Unsplash · Chase Chappell · Unsplash License", "creditUrl": "https://unsplash.com/photos/a-man-writing-on-a-white-board-with-orange-markers--kVD3jtpumo", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-one-pilot.webp" }, "tags": [ "implementation", "pilot" ], "curatedDistractors": [ "rollout", "policy", "audit" ], "uid": "fhes9nezz5t9" }, { "id": "aisb-fact-baseline", "shape": "fact", "title": "Measure Before and After", "body": "Record current time, cost, error rate, turnaround, and customer impact before changing the workflow. Without a baseline, improvement becomes guesswork.", "illustration": { "kind": "photo", "imagePrompt": "A small business owner studies a dashboard of charts and metrics on a laptop, comparing before-and-after performance numbers.", "imageSearchTerm": "business owner reviewing performance dashboard", "alt": "A business owner comparing before-and-after workflow results", "depictable": true, "credit": "Unsplash · Campaign Creators · Unsplash License", "creditUrl": "https://unsplash.com/photos/person-using-macbook-pro-pypeCEaJeZY", "subject": "Over-the-shoulder view of a person typing on a MacBook whose screen shows a web-analytics dashboard (pie chart, two line sparklines, horizontal bar rows) on a white desk with glasses and a phone; text on screen too small to read.", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-baseline.webp" }, "tags": [ "implementation", "measurement" ], "uid": "1wd5tu1qz1nnd" }, { "id": "aisb-fact-acceptance-rate", "shape": "fact", "title": "Track Usable Output", "body": "A useful pilot metric is the share of AI outputs accepted after review, along with the time and type of corrections required.", "tags": [ "implementation", "measurement" ], "illustration": { "imageSearchTerm": "reviewing document making corrections", "imagePrompt": "An office worker reviews a printed report at a desk, marking corrections in red pen beside an open laptop.", "alt": "Track Usable Output", "depictable": true }, "uid": "ctamvm1lohiz6" }, { "id": "aisb-concept-task-inventory", "shape": "concept", "conceptKind": "thing", "name": "Task inventory", "clues": [ "A list used before selecting an AI tool.", "It focuses attention on recurring work and bottlenecks.", "It records frequency, time, structure, and error consequence for candidate workflows." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "10mlbotmfm6n" }, { "id": "aisb-concept-baseline", "shape": "concept", "conceptKind": "thing", "name": "Baseline", "clues": [ "A reference point established before a change.", "It prevents teams from relying only on impressions of improvement.", "It records current time, cost, errors, or outcomes for comparison with a pilot." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "16378zl1mbbzx5" }, { "id": "aisb-concept-pilot", "shape": "concept", "conceptKind": "thing", "name": "Pilot", "clues": [ "A limited implementation designed to produce evidence.", "It reduces exposure while a workflow is tested and refined.", "It uses a defined scope, owner, period, and success criteria before wider rollout." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "10c1ghq13vfz4m" }, { "id": "aisb-concept-parallel-run", "shape": "concept", "conceptKind": "thing", "name": "Parallel run", "clues": [ "A transition technique that preserves an existing process temporarily.", "It makes side-by-side quality and timing comparisons possible.", "The new AI-assisted workflow operates alongside the manual workflow before replacement." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "1jl62vu1d6i4xc" }, { "id": "aisb-concept-acceptance-criteria", "shape": "concept", "conceptKind": "thing", "name": "Acceptance criteria", "clues": [ "A quality tool defined before reviewing results.", "It makes approval less dependent on mood or impressive wording.", "It lists the conditions an output must satisfy to be considered usable." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "1m08x527xv40o" }, { "id": "aisb-concept-roi", "shape": "concept", "conceptKind": "thing", "name": "Return on investment", "clues": [ "A comparison between value received and resources spent.", "For a pilot, its inputs can include subscriptions, setup, review time, savings, and avoided cost.", "It helps decide whether to stop, adjust, or expand an AI workflow." ], "tags": [ "implementation", "pilot", "key-concept" ], "uid": "m73mabgk5uvb" }, { "id": "aisb-mcq-pilot-success", "shape": "mcq", "prompt": { "modality": "text", "value": "When should a small business expand an AI pilot?" }, "options": [ { "modality": "text", "value": "As soon as the output looks impressive", "short": "When output looks impressive" }, { "modality": "text", "value": "When measured results meet criteria and risks are controlled", "short": "When results & risks check out" }, { "modality": "text", "value": "When competitors mention the same tool", "short": "When rivals mention the tool" }, { "modality": "text", "value": "After buying an annual plan", "short": "After buying an annual plan" } ], "correctIndex": 1, "explanation": "Expansion should follow evidence and control, not novelty or sunk cost.", "tags": [ "implementation", "pilot" ], "uid": "xliwnk1c5ywlo" }, { "id": "aisb-fact-policy-light", "shape": "fact", "title": "Keep the First Policy Short", "body": "A beginner AI policy can fit on one page: approved tools, prohibited data, review requirements, external-use rules, incident reporting, and an owner.", "illustration": { "kind": "photo", "imagePrompt": "Candid editorial photograph of a small-business owner holding up a single sheet of paper edge-on while explaining a concise policy to two employees, the paper face turned completely away from the camera, practical team discussion, no visible writing, words, letters, numbers, screens, or logos.", "imageSearchTerm": "small business owner reviewing one page policy", "alt": "A business owner reviewing a concise one-page policy", "credit": "AI-generated (gpt-image-1.5)", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-policy-light.webp" }, "tags": [ "implementation", "policy" ], "uid": "1f8qg2f1ntxsyd" }, { "id": "aisb-fact-owner", "shape": "fact", "title": "Every Workflow Needs an Owner", "body": "Someone must maintain instructions, permissions, source material, tests, and performance—even when the workflow runs automatically.", "tags": [ "implementation", "governance" ], "illustration": { "imageSearchTerm": "automated workflow maintenance concept", "imagePrompt": "A single glowing key holds together a network of interconnected gears and documents, representing one person responsible for maintaining a system.", "alt": "Every Workflow Needs an Owner", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:Lick_Observatory_APF.jpg", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-owner.webp" }, "uid": "2nr07b1gc8je5" }, { "id": "aisb-fact-quarterly-review", "shape": "fact", "title": "Review the System, Not Just the Output", "body": "Periodically review tool access, vendor terms, failed outputs, time savings, source documents, prompts, and whether the workflow still solves the right problem.", "tags": [ "implementation", "governance" ], "illustration": { "imageSearchTerm": "periodic business system review", "imagePrompt": "A magnifying glass hovers over a checklist of interconnected icons for tools, contracts, and documents, symbolizing a periodic system review.", "alt": "Review the System, Not Just the Output", "depictable": false, "credit": "Wikimedia Commons · see source", "creditUrl": "https://commons.wikimedia.org/wiki/File:The_Ring_Of_Periodic_Elements_(TROPE).png", "url": "https://cdn.recurxive.com/packs/ai-for-small-business/images/aisb-fact-quarterly-review.webp" }, "uid": "mgfpeef24yma" }, { "id": "aisb-concept-ai-policy", "shape": "concept", "conceptKind": "thing", "name": "AI use policy", "clues": [ "A governance document sized to the business.", "It gives staff common boundaries before a problem occurs.", "It names approved tools, prohibited data, review rules, external-use expectations, and incident reporting." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "l0lx7v1dwm8t5" }, { "id": "aisb-concept-prompt-library", "shape": "concept", "conceptKind": "thing", "name": "Prompt library", "clues": [ "A reusable knowledge asset rather than a single conversation.", "It helps a team repeat successful AI-assisted work.", "It stores tested instructions, examples, source materials, and review checklists." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "f5bci71cvh4wb" }, { "id": "aisb-concept-approval-gate", "shape": "concept", "conceptKind": "thing", "name": "Approval gate", "clues": [ "A control placed before consequential action.", "It limits autonomy without preventing drafting or analysis.", "It requires a person to confirm an action before money moves, a message sends, or a record changes." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "ddnhyn19v3hz1" }, { "id": "aisb-concept-audit-log", "shape": "concept", "conceptKind": "thing", "name": "Audit log", "clues": [ "A record that supports troubleshooting and accountability.", "It helps reconstruct what a workflow did and why review failed.", "It records inputs, outputs, actions, timestamps, errors, and approvals." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "191k53mgwilh0" }, { "id": "aisb-concept-escalation-rule", "shape": "concept", "conceptKind": "thing", "name": "Escalation rule", "clues": [ "A boundary for uncertainty, exceptions, or authority.", "It prevents a system from bluffing its way through unusual cases.", "It defines when automation stops and a case transfers to a responsible person." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "zg2g1uelw5ng" }, { "id": "aisb-concept-kill-switch", "shape": "concept", "conceptKind": "thing", "name": "Kill switch", "clues": [ "An operational safeguard for a malfunctioning workflow.", "It should be quick to use and known to responsible staff.", "It immediately stops automated actions while an incident is investigated." ], "tags": [ "implementation", "governance", "key-concept" ], "uid": "1pmu91r1gu4eop" }, { "id": "aisb-mcq-operating-system", "shape": "mcq", "prompt": { "modality": "text", "value": "What turns an AI experiment into a dependable business capability?" }, "options": [ { "modality": "text", "value": "More subscriptions", "short": "Buying more subscriptions" }, { "modality": "text", "value": "A famous prompt", "short": "Using a famous prompt" }, { "modality": "text", "value": "Removing people from every decision", "short": "Remove people from decisions" }, { "modality": "text", "value": "Owned workflows with controls, documentation, measurement, and review", "short": "Owned workflows with controls" } ], "correctIndex": 3, "explanation": "Dependability comes from the operating system around the model, not the model alone.", "tags": [ "implementation", "governance" ], "uid": "1slu54t9kkebb" }, { "id": "czr-ai-for-small-business-aisb-def-hallucination", "shape": "cloze", "tags": [ "foundations", "reliability" ], "template": "___ — A confident-sounding AI output that contains invented, unsupported, or incorrect information", "answer": "Hallucination", "distractors": [ "Grounded response", "Verified fact", "Source citation" ], "derivedFrom": "aisb-def-hallucination", "uid": "tsibtu4du4c0" }, { "id": "czr-ai-for-small-business-aisb-def-workflow", "shape": "cloze", "tags": [ "foundations", "workflow" ], "template": "___ — A repeatable sequence of inputs, decisions, actions, and outputs used to complete a business task", "answer": "Workflow", "distractors": [ "Project", "Policy", "Single task" ], "derivedFrom": "aisb-def-workflow", "uid": "5a0r4x1b9xtlt" }, { "id": "czr-ai-for-small-business-aisb-def-acceptance-criteria", "shape": "cloze", "tags": [ "prompting", "quality" ], "template": "___ — Specific conditions an output must satisfy before it is considered usable", "answer": "Acceptance criteria", "distractors": [ "Business context", "Source material", "First draft" ], "derivedFrom": "aisb-def-acceptance-criteria", "uid": "14kvasv13iy3lp" }, { "id": "czr-ai-for-small-business-aisb-def-prompt-library", "shape": "cloze", "tags": [ "prompting", "reuse" ], "template": "___ — A maintained collection of tested instructions, examples, and checklists for recurring work", "answer": "Prompt library", "distractors": [ "Brand guide", "Knowledge base", "Audit log" ], "derivedFrom": "aisb-def-prompt-library", "explanation": "Compare: Audit log — A record that supports troubleshooting and accountability.", "uid": "1iyx4bw1hqdb5a" }, { "id": "czr-ai-for-small-business-aisb-def-triangulation", "shape": "cloze", "tags": [ "research", "verification" ], "template": "___ — Checking a material claim against multiple independent sources or methods before relying on it", "answer": "Triangulation", "distractors": [ "Summarization", "Brainstorming", "Forecasting" ], "derivedFrom": "aisb-def-triangulation", "uid": "wfqmc9mpm9nl" }, { "id": "czr-ai-for-small-business-aisb-def-sensitivity-analysis", "shape": "cloze", "tags": [ "analysis", "forecasting" ], "template": "___ — Testing how a result changes when an important assumption or input changes", "answer": "Sensitivity analysis", "distractors": [ "Trend analysis", "Data cleaning", "Reconciliation" ], "derivedFrom": "aisb-def-sensitivity-analysis", "uid": "jll0id16jke9f" }, { "id": "czr-ai-for-small-business-aisb-def-content-atom", "shape": "cloze", "tags": [ "marketing", "content" ], "template": "___ — A substantial source asset that can be adapted into multiple smaller channel-specific pieces", "answer": "Content atom", "distractors": [ "Brand brief", "Proof point", "Editorial calendar" ], "derivedFrom": "aisb-def-content-atom", "uid": "9pe3nr126cvcp" }, { "id": "czr-ai-for-small-business-aisb-def-escalation-rule", "shape": "cloze", "tags": [ "customer-service", "escalation" ], "template": "___ — A defined condition that stops automation and transfers a case to human review", "answer": "Escalation rule", "distractors": [ "Approval gate", "Exception path", "Kill switch" ], "derivedFrom": "aisb-def-escalation-rule", "explanation": "Compare: Approval gate — A required human confirmation before a workflow performs a consequential action. Kill switch — An operational safeguard for a malfunctioning workflow.", "uid": "lm0ua21ng410o" }, { "id": "czr-ai-for-small-business-aisb-def-sop", "shape": "cloze", "tags": [ "operations", "sop" ], "template": "___ — A maintained description of how to perform a recurring task, including steps, responsibilities, controls, and exceptions", "answer": "Standard operating procedure", "distractors": [ "Policy", "Decision memo", "Project brief" ], "derivedFrom": "aisb-def-sop", "uid": "75aaqc19wvuj0" }, { "id": "czr-ai-for-small-business-aisb-def-approval-gate", "shape": "cloze", "tags": [ "automation", "governance" ], "template": "___ — A required human confirmation before a workflow performs a consequential action", "answer": "Approval gate", "distractors": [ "Audit log", "Escalation rule", "Kill switch" ], "derivedFrom": "aisb-def-approval-gate", "explanation": "Compare: Audit log — A record that supports troubleshooting and accountability. Escalation rule — A defined condition that stops automation and transfers a case to human review.", "uid": "1i2q89a1908dse" }, { "id": "czr-ai-for-small-business-aisb-def-material-term", "shape": "cloze", "tags": [ "high-stakes", "contracts" ], "template": "___ — A contract provision important enough that it could affect cost, rights, obligations, risk, or the decision to sign", "answer": "Material term", "distractors": [ "Formatting note", "Nonbinding example", "Document heading" ], "derivedFrom": "aisb-def-material-term", "uid": "mod34eu15gae" }, { "id": "czr-ai-for-small-business-aisb-def-adverse-impact", "shape": "cloze", "tags": [ "people", "fairness" ], "template": "___ — A substantially different negative outcome for a protected group caused by a seemingly neutral employment practice", "answer": "Adverse impact", "distractors": [ "Reasonable accommodation", "Objective criterion", "Individual review" ], "derivedFrom": "aisb-def-adverse-impact", "uid": "1lec1og1e5cfk" }, { "id": "czr-ai-for-small-business-aisb-def-data-retention", "shape": "cloze", "tags": [ "privacy", "data" ], "template": "___ — How long a service or business keeps information before deletion or archival", "answer": "Data retention", "distractors": [ "Data minimization", "Access control", "Encryption" ], "derivedFrom": "aisb-def-data-retention", "uid": "9ocj7kga70nk" }, { "id": "czr-ai-for-small-business-aisb-def-human-in-loop", "shape": "cloze", "tags": [ "governance", "oversight" ], "template": "___ — A workflow design in which a person reviews, approves, corrects, or can stop AI-supported work at a meaningful point", "answer": "Human in the loop", "distractors": [ "Audit log", "Kill switch", "Automated trigger" ], "derivedFrom": "aisb-def-human-in-loop", "explanation": "Compare: Audit log — A record that supports troubleshooting and accountability. Kill switch — An operational safeguard for a malfunctioning workflow.", "uid": "1bmqv5jrbwe57" }, { "id": "cz-ai-for-small-business-aisb-fact-copyright", "shape": "cloze", "tags": [ "marketing", "copyright" ], "template": "The ___ says AI-assisted work can qualify for protection when humans contribute sufficient expressive authorship, but prompts alone generally do not make a purely AI-generated output copyrightable.", "answer": "U.S. Copyright Office", "distractors": [ "Federal Trade Commission", "Small Business Administration", "National Institute of Standards and Technology" ], "derivedFrom": "aisb-fact-copyright", "uid": "1y0tvrcgz40ao" }, { "id": "cz-ai-for-small-business-aisb-fact-business-data", "shape": "cloze", "tags": [ "privacy", "vendors" ], "template": "For example, OpenAI states that ___, Enterprise, and API data are not used to train models by default. Every vendor should be evaluated separately before sensitive use.", "answer": "ChatGPT Business", "distractors": [ "ChatGPT Free", "ChatGPT Plus", "ChatGPT Pro" ], "derivedFrom": "aisb-fact-business-data", "uid": "1jx2o5hfl9l17" }, { "id": "cz-ai-for-small-business-aisb-fact-one-pilot", "shape": "cloze", "tags": [ "implementation", "pilot" ], "template": "A focused 30-day ___ creates evidence. Tool hopping creates opinions without a stable process or baseline.", "answer": "pilot", "distractors": [ "rollout", "policy", "audit" ], "derivedFrom": "aisb-fact-one-pilot", "uid": "1s1e2bnju92ln" }, { "id": "tf-d-ai-for-small-business-aisb-def-hallucination", "shape": "trueFalse", "tags": [ "foundations", "reliability" ], "statement": "Hallucination: A confident-sounding AI output that contains invented, unsupported, or incorrect information.", "isTrue": true, "why": "That definition is correct for \"Hallucination\".", "derivedFrom": "aisb-def-hallucination", "uid": "1d97rwzr07srt" }, { "id": "tf-d-ai-for-small-business-aisb-def-workflow", "shape": "trueFalse", "tags": [ "foundations", "workflow" ], "statement": "Workflow: A repeatable sequence of inputs, decisions, actions, and outputs used to complete a business task.", "isTrue": true, "why": "That definition is correct for \"Workflow\".", "derivedFrom": "aisb-def-workflow", "uid": "175dnxw1w2swig" }, { "id": "tf-d-ai-for-small-business-aisb-def-acceptance-criteria", "shape": "trueFalse", "tags": [ "prompting", "quality" ], "statement": "Acceptance criteria: Specific conditions an output must satisfy before it is considered usable.", "isTrue": true, "why": "That definition is correct for \"Acceptance criteria\".", "derivedFrom": "aisb-def-acceptance-criteria", "uid": "1wy44bc1b2m9ty" }, { "id": "tf-df-ai-for-small-business-aisb-def-prompt-library", "shape": "trueFalse", "tags": [ "prompting", "reuse" ], "statement": "Acceptance criteria: A maintained collection of tested instructions, examples, and checklists for recurring work.", "isTrue": false, "why": "That's the definition of \"Prompt library\", not \"Acceptance criteria\".", "derivedFrom": "aisb-def-prompt-library", "uid": "1rslvtvj2m0j5" }, { "id": "tf-d-ai-for-small-business-aisb-def-triangulation", "shape": "trueFalse", "tags": [ "research", "verification" ], "statement": "Triangulation: Checking a material claim against multiple independent sources or methods before relying on it.", "isTrue": true, "why": "That definition is correct for \"Triangulation\".", "derivedFrom": "aisb-def-triangulation", "uid": "1y9mqzkwlyvb4" }, { "id": "tf-df-ai-for-small-business-aisb-def-sensitivity-analysis", "shape": "trueFalse", "tags": [ "analysis", "forecasting" ], "statement": "Escalation rule: Testing how a result changes when an important assumption or input changes.", "isTrue": false, "why": "That's the definition of \"Sensitivity analysis\", not \"Escalation rule\".", "derivedFrom": "aisb-def-sensitivity-analysis", "uid": "5trj4pq7mshh" }, { "id": "tf-d-ai-for-small-business-aisb-def-content-atom", "shape": "trueFalse", "tags": [ "marketing", "content" ], "statement": "Content atom: A substantial source asset that can be adapted into multiple smaller channel-specific pieces.", "isTrue": true, "why": "That definition is correct for \"Content atom\".", "derivedFrom": "aisb-def-content-atom", "uid": "bqc638a6xgm" }, { "id": "tf-df-ai-for-small-business-aisb-def-escalation-rule", "shape": "trueFalse", "tags": [ "customer-service", "escalation" ], "statement": "Sensitivity analysis: A defined condition that stops automation and transfers a case to human review.", "isTrue": false, "why": "That's the definition of \"Escalation rule\", not \"Sensitivity analysis\".", "derivedFrom": "aisb-def-escalation-rule", "uid": "1dh8jbsw22jac" }, { "id": "tf-df-ai-for-small-business-aisb-def-sop", "shape": "trueFalse", "tags": [ "operations", "sop" ], "statement": "Data retention: A maintained description of how to perform a recurring task, including steps, responsibilities, controls, and exceptions.", "isTrue": false, "why": "That's the definition of \"Standard operating procedure\", not \"Data retention\".", "derivedFrom": "aisb-def-sop", "uid": "vy2em91i63h5b" }, { "id": "tf-d-ai-for-small-business-aisb-def-approval-gate", "shape": "trueFalse", "tags": [ "automation", "governance" ], "statement": "Approval gate: A required human confirmation before a workflow performs a consequential action.", "isTrue": true, "why": "That definition is correct for \"Approval gate\".", "derivedFrom": "aisb-def-approval-gate", "uid": "sqny6x4nqkut" }, { "id": "tf-df-ai-for-small-business-aisb-def-material-term", "shape": "trueFalse", "tags": [ "high-stakes", "contracts" ], "statement": "Triangulation: A contract provision important enough that it could affect cost, rights, obligations, risk, or the decision to sign.", "isTrue": false, "why": "That's the definition of \"Material term\", not \"Triangulation\".", "derivedFrom": "aisb-def-material-term", "uid": "sy657s1luvpf0" }, { "id": "tf-d-ai-for-small-business-aisb-def-adverse-impact", "shape": "trueFalse", "tags": [ "people", "fairness" ], "statement": "Adverse impact: A substantially different negative outcome for a protected group caused by a seemingly neutral employment practice.", "isTrue": true, "why": "That definition is correct for \"Adverse impact\".", "derivedFrom": "aisb-def-adverse-impact", "uid": "1ea92u1lgj8qt" }, { "id": "tf-d-ai-for-small-business-aisb-def-data-retention", "shape": "trueFalse", "tags": [ "privacy", "data" ], "statement": "Data retention: How long a service or business keeps information before deletion or archival.", "isTrue": true, "why": "That definition is correct for \"Data retention\".", "derivedFrom": "aisb-def-data-retention", "uid": "144st9pxss2sp" }, { "id": "tf-df-ai-for-small-business-aisb-def-human-in-loop", "shape": "trueFalse", "tags": [ "governance", "oversight" ], "statement": "Approval gate: A workflow design in which a person reviews, approves, corrects, or can stop AI-supported work at a meaningful point.", "isTrue": false, "why": "That's the definition of \"Human in the loop\", not \"Approval gate\".", "derivedFrom": "aisb-def-human-in-loop", "uid": "rp4tny1f3g490" }, { "id": "ot-capabilities-hallucination", "shape": "mcq", "tags": [ "hallucination", "reliability" ], "prompt": { "modality": "text", "value": "In AI terms, what is a 'hallucination'?" }, "options": [ { "modality": "text", "value": "A slowdown caused by too many simultaneous requests", "short": "Slowdown from too many requests" }, { "modality": "text", "value": "A confident-sounding output that contains invented or unsupported information", "short": "Confident but invented info" }, { "modality": "text", "value": "A deliberate refusal to answer a sensitive question", "short": "Refusal to answer a question" }, { "modality": "text", "value": "A setting that shortens the length of a response", "short": "Setting that shortens a reply" } ], "correctIndex": 1, "explanation": "A hallucination is a confident-sounding AI output containing invented, unsupported, or incorrect information. The other options describe unrelated behaviors, not fabricated content.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "14wv42zdaqkkz" }, { "id": "ot-capabilities-analysis", "shape": "mcq", "tags": [ "analysis", "generation" ], "prompt": { "modality": "text", "value": "Which task is an example of AI analysis rather than generation?" }, "options": [ { "modality": "text", "value": "Writing a customer reply from your notes", "short": "Write a customer reply" }, { "modality": "text", "value": "Brainstorming names for a new promotion", "short": "Brainstorm promotion names" }, { "modality": "text", "value": "Finding patterns in a batch of customer reviews you upload", "short": "Find patterns in reviews" }, { "modality": "text", "value": "Drafting a job description from a short brief", "short": "Draft a job description" } ], "correctIndex": 2, "explanation": "Analysis finds patterns or implications in material you provide, like uploaded reviews. Drafting, writing, and brainstorming all create new drafts, which is generation.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "i3pt2xupygsv" }, { "id": "ot-capabilities-fluency-not", "shape": "mcq", "tags": [ "fluency", "verification" ], "prompt": { "modality": "text", "value": "Which statement about a polished AI answer is NOT correct?" }, "options": [ { "modality": "text", "value": "A polished answer can still be incomplete or stale", "short": "Polished answers can be stale" }, { "modality": "text", "value": "Material claims should be verified before acting", "short": "Verify material claims first" }, { "modality": "text", "value": "Fluent, confident wording confirms the claim is accurate", "short": "Fluent wording confirms accuracy" }, { "modality": "text", "value": "The same request can produce different answers", "short": "Same request, different answers" } ], "correctIndex": 2, "explanation": "Fluency is not evidence of accuracy, so that statement is false. A polished answer can be wrong or stale, outputs vary, and claims should be verified.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1t8pnpj8sy4c7" }, { "id": "ot-capabilities-stats", "shape": "mcq", "tags": [ "evidence", "adoption-stats" ], "prompt": { "modality": "text", "value": "Why can't the Census Bureau's roughly 17 to 20 percent figure and the Chamber's 58 percent figure be treated as interchangeable?" }, "options": [ { "modality": "text", "value": "One counted only failed AI projects", "short": "One counted only failed projects" }, { "modality": "text", "value": "They measured entirely different decades", "short": "They measured different decades" }, { "modality": "text", "value": "Both figures were later retracted as errors", "short": "Both were retracted as errors" }, { "modality": "text", "value": "They used different populations, definitions, and methods", "short": "Different populations & methods" } ], "correctIndex": 3, "explanation": "The figures differ because they surveyed different populations with different definitions and methods. Always ask who was surveyed and what counted as use before repeating a headline.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1m8jept1yk7esb" }, { "id": "ot-workflow-definition", "shape": "mcq", "tags": [ "workflow", "definition" ], "prompt": { "modality": "text", "value": "How does this lesson define a 'workflow'?" }, "options": [ { "modality": "text", "value": "A one-time decision about which software to buy", "short": "One-time software-buying choice" }, { "modality": "text", "value": "A ranked list of AI vendors to compare", "short": "Ranked list of AI vendors" }, { "modality": "text", "value": "The total monthly cost of an AI subscription", "short": "Monthly subscription cost" }, { "modality": "text", "value": "A repeatable sequence of inputs, decisions, actions, and outputs", "short": "Repeatable sequence of steps" } ], "correctIndex": 3, "explanation": "A workflow is a repeatable sequence of inputs, decisions, actions, and outputs used to complete a business task. The other options describe purchases or costs, not the process itself.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "dhv113qovio3" }, { "id": "ot-workflow-worst", "shape": "mcq", "tags": [ "high-stakes", "task-fit" ], "prompt": { "modality": "text", "value": "Which task is the WORST candidate for a first AI pilot?" }, "options": [ { "modality": "text", "value": "Turning meeting notes into a checklist", "short": "Turn notes into a checklist" }, { "modality": "text", "value": "Deciding whether to fire an employee", "short": "Decide whether to fire someone" }, { "modality": "text", "value": "Summarizing an internal meeting", "short": "Summarize an internal meeting" }, { "modality": "text", "value": "Drafting routine customer replies", "short": "Draft routine customer replies" } ], "correctIndex": 1, "explanation": "Firing an employee has ambiguous inputs, legal consequences, and an irreversible decision, making it a poor first pilot. The others are frequent, reviewable, and low consequence.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "19iaq4f4k3h2h" }, { "id": "ot-workflow-augment", "shape": "mcq", "tags": [ "augment", "automation" ], "prompt": { "modality": "text", "value": "What is the main benefit of running AI alongside your existing process before automating it?" }, "options": [ { "modality": "text", "value": "It guarantees effortless passive income", "short": "Guarantees passive income" }, { "modality": "text", "value": "It removes the need for any owner review", "short": "Removes need for owner review" }, { "modality": "text", "value": "It exposes errors, exceptions, and missing instructions", "short": "Exposes errors and gaps" }, { "modality": "text", "value": "It eliminates all future software costs", "short": "Eliminates future software costs" } ], "correctIndex": 2, "explanation": "A parallel run reveals errors, exceptions, and missing instructions that a generic demo never shows. It does not remove review or guarantee income and savings.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1n7izam11slqwe" }, { "id": "ot-workflow-best-task", "shape": "mcq", "tags": [ "best-first-task", "task-fit" ], "prompt": { "modality": "text", "value": "According to the lesson, a strong first AI task should be:" }, "options": [ { "modality": "text", "value": "Frequent, easy to check, and low consequence if wrong", "short": "Frequent, checkable, low-stakes" }, { "modality": "text", "value": "Rare, ambiguous, and legally binding", "short": "Rare, ambiguous, legally binding" }, { "modality": "text", "value": "Complex and requiring expert judgment", "short": "Complex, needs expert judgment" }, { "modality": "text", "value": "Irreversible but high in value", "short": "Irreversible but high-value" } ], "correctIndex": 0, "explanation": "A good first task is frequent, structured, easy to review, and low consequence when the first draft is wrong. The other traits describe high-risk, poor-fit candidates.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "emm6ru1kqdus" }, { "id": "ot-brief-ingredients", "shape": "mcq", "tags": [ "brief", "prompt-structure" ], "prompt": { "modality": "text", "value": "A useful AI request usually contains which five ingredients?" }, "options": [ { "modality": "text", "value": "Greeting, body, closing, subject line, and attachment", "short": "Greeting, body, subject, file" }, { "modality": "text", "value": "Model, temperature, token limit, seed, and format", "short": "Model, temp, tokens, seed" }, { "modality": "text", "value": "Task, context, audience, constraints, and expected output", "short": "Task, context, audience, output" }, { "modality": "text", "value": "Keywords, tone, length, deadline, and signature", "short": "Keywords, tone, length, deadline" } ], "correctIndex": 2, "explanation": "The guide defines a good prompt as a short business brief with the task, relevant context, the audience, constraints, and the expected output.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "x4nzofqv09z3" }, { "id": "ot-brief-acceptance-def", "shape": "mcq", "tags": [ "acceptance-criteria", "definitions" ], "prompt": { "modality": "text", "value": "Specific conditions an output must satisfy before it is considered usable are called:" }, "options": [ { "modality": "text", "value": "Context" }, { "modality": "text", "value": "A prompt library" }, { "modality": "text", "value": "The draft pass" }, { "modality": "text", "value": "Acceptance criteria" } ], "correctIndex": 3, "explanation": "Acceptance criteria are the specific conditions an output must meet to be usable; a prompt library and draft pass are separate concepts.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1hs1t26bhlel2" }, { "id": "ot-brief-examples", "shape": "mcq", "tags": [ "examples", "voice" ], "prompt": { "modality": "text", "value": "Why include an approved example email in a brief instead of only saying 'sound professional'?" }, "options": [ { "modality": "text", "value": "It lets the AI skip reading the constraints", "short": "Lets AI skip the constraints" }, { "modality": "text", "value": "It guarantees the output needs no human review", "short": "Output needs no human review" }, { "modality": "text", "value": "It uses fewer tokens than writing adjectives", "short": "Uses fewer tokens than words" }, { "modality": "text", "value": "It shows rhythm, warmth, terminology, and level of detail better than adjectives", "short": "Shows tone & detail concretely" } ], "correctIndex": 3, "explanation": "Examples communicate voice and detail more concretely than adjectives. They do not remove the need for review, and sensitive details should be stripped first.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "7fx1i517jq6wx" }, { "id": "ot-brief-criteria-not", "shape": "mcq", "tags": [ "acceptance-criteria", "review" ], "prompt": { "modality": "text", "value": "Acceptance criteria might specify all of the following EXCEPT:" }, "options": [ { "modality": "text", "value": "A maximum length or specified format", "short": "Maximum length or format" }, { "modality": "text", "value": "A policy that must be followed", "short": "A policy that must be followed" }, { "modality": "text", "value": "The model's temperature and random seed", "short": "Model temperature and seed" }, { "modality": "text", "value": "Required facts the output must include", "short": "Required facts to include" } ], "correctIndex": 2, "explanation": "Criteria cover required facts, length, format, prohibited claims, and policies. Model temperature and seed are technical settings, not acceptance criteria from the guide.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "b20gz67sb0gq" }, { "id": "ot-iter-library-def", "shape": "mcq", "tags": [ "prompt-library", "definitions" ], "prompt": { "modality": "text", "value": "A maintained collection of tested instructions, examples, and checklists for recurring work is called a:" }, "options": [ { "modality": "text", "value": "Critique pass" }, { "modality": "text", "value": "Acceptance criterion" }, { "modality": "text", "value": "Prompt library" }, { "modality": "text", "value": "Draft pass" } ], "correctIndex": 2, "explanation": "A prompt library stores tested instructions, examples, and checklists for reuse. The other terms name single steps or conditions, not the stored collection.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "90yjt0jqb7xs" }, { "id": "ot-iter-critique-pass", "shape": "mcq", "tags": [ "critique", "iteration" ], "prompt": { "modality": "text", "value": "During the critique pass of an AI conversation, what should you do?" }, "options": [ { "modality": "text", "value": "Identify weaknesses against explicit criteria", "short": "Find weaknesses vs criteria" }, { "modality": "text", "value": "Publish the output to save time", "short": "Publish it to save time" }, { "modality": "text", "value": "Delete the draft and start over", "short": "Delete and start over" }, { "modality": "text", "value": "Create the first workable version", "short": "Create the first version" } ], "correctIndex": 0, "explanation": "The critique pass checks the draft against explicit criteria for weaknesses and unsupported claims. Creating the first workable version is the draft pass.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "wi40zb186vpcx" }, { "id": "ot-iter-save-pattern", "shape": "mcq", "tags": [ "prompt-library", "reuse" ], "prompt": { "modality": "text", "value": "When a workflow succeeds, what is most valuable to save?" }, "options": [ { "modality": "text", "value": "The AI's temperature setting", "short": "The AI temperature setting" }, { "modality": "text", "value": "The instructions, source material, examples, checklist, and owner", "short": "Inputs, examples, checklist" }, { "modality": "text", "value": "Only the final polished draft", "short": "Only the final draft" }, { "modality": "text", "value": "A screenshot of the chat window", "short": "A screenshot of the chat" } ], "correctIndex": 1, "explanation": "Saving the full pattern (instructions, sources, examples, checklist, owner, and date) is more reusable than keeping just one finished output.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "14gse6d1emg14f" }, { "id": "ot-iter-diagnose-promise", "shape": "mcq", "tags": [ "diagnosis", "constraints" ], "prompt": { "modality": "text", "value": "A draft makes an unsafe promise to a customer. Using the diagnosis approach, what was most likely incomplete?" }, "options": [ { "modality": "text", "value": "The approved example emails", "short": "The example emails" }, { "modality": "text", "value": "The constraints", "short": "The constraints" }, { "modality": "text", "value": "The expected output format", "short": "The output format" }, { "modality": "text", "value": "The maximum length", "short": "The maximum length" } ], "correctIndex": 1, "explanation": "An unsafe promise signals incomplete constraints. A wrong fact points to missing sources, generic language to vague examples, and bad structure to an undefined format.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "10puqhj1hhtu5x" }, { "id": "ot-research-primary", "shape": "mcq", "tags": [ "primary-sources", "research" ], "prompt": { "modality": "text", "value": "Which is the strongest example of a primary source for a material business decision?" }, "options": [ { "modality": "text", "value": "A news article quoting an industry analyst", "short": "News article quoting an analyst" }, { "modality": "text", "value": "A blog post summarizing a government report", "short": "Blog post summarizing a report" }, { "modality": "text", "value": "An original customer survey your business conducted", "short": "An original customer survey" }, { "modality": "text", "value": "A competitor's marketing page describing a trend", "short": "Competitor's marketing page" } ], "correctIndex": 2, "explanation": "A primary source is original data or firsthand evidence. Your own survey is firsthand; blogs, news articles, and marketing pages are secondary summaries or positioning.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "16oeniv1rib04v" }, { "id": "ot-research-triangulate", "shape": "mcq", "tags": [ "triangulation", "research" ], "prompt": { "modality": "text", "value": "Which action best demonstrates triangulation when validating a demand claim?" }, "options": [ { "modality": "text", "value": "Comparing customer interviews, transaction data, and public statistics", "short": "Interviews, data, and stats" }, { "modality": "text", "value": "Averaging two guesses produced by the same model", "short": "Average two guesses, same model" }, { "modality": "text", "value": "Asking the AI to restate its estimate more confidently", "short": "AI restates estimate confidently" }, { "modality": "text", "value": "Reading three articles that all cite the same original study", "short": "Three articles citing one study" } ], "correctIndex": 0, "explanation": "Triangulation checks a claim against multiple independent methods. Articles repeating one study, or one model's own guesses, are not independent.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "185la0ovfwmjm" }, { "id": "ot-research-when-search", "shape": "mcq", "tags": [ "current-search", "research" ], "prompt": { "modality": "text", "value": "Which question is LEAST likely to require an AI mode that searches the web for current sources?" }, "options": [ { "modality": "text", "value": "A new product's latest features", "short": "New product's latest features" }, { "modality": "text", "value": "The meaning of a standard arithmetic operation", "short": "Standard arithmetic meaning" }, { "modality": "text", "value": "The current interest rate on a business loan", "short": "Current business-loan rate" }, { "modality": "text", "value": "A competitor's current pricing", "short": "Competitor's current pricing" } ], "correctIndex": 1, "explanation": "Current search matters when the answer could have changed: prices, rates, and features. A fixed arithmetic definition is stable and does not need fresh sources.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1jy7lcw1fmkuei" }, { "id": "ot-research-definitions", "shape": "mcq", "tags": [ "definitions", "survey-data" ], "prompt": { "modality": "text", "value": "The U.S. Census found about 17 to 20% of businesses using AI, while a Chamber survey reported 58%. Why do these differ so much?" }, "options": [ { "modality": "text", "value": "One of the two figures is simply a typo", "short": "One figure is simply a typo" }, { "modality": "text", "value": "Adoption collapsed sharply between the two surveys", "short": "Adoption collapsed between them" }, { "modality": "text", "value": "The surveys were conducted in different countries", "short": "Different countries surveyed" }, { "modality": "text", "value": "The surveys asked different populations and defined using AI differently", "short": "Different populations & AI defs" } ], "correctIndex": 3, "explanation": "Estimates diverge because surveys sample different populations and define 'using AI' differently, so the definition changes the number.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "19g5cit1t8hjij" }, { "id": "ot-data-inference", "shape": "mcq", "tags": [ "observation-inference", "data" ], "prompt": { "modality": "text", "value": "Which statement is an inference rather than an observation?" }, "options": [ { "modality": "text", "value": "Total revenue was lower than in May", "short": "Revenue lower than in May" }, { "modality": "text", "value": "The new supplier caused the increase in returns", "short": "New supplier caused returns" }, { "modality": "text", "value": "Returns increased in June", "short": "Returns increased in June" }, { "modality": "text", "value": "Average order value fell in June", "short": "Avg order value fell in June" } ], "correctIndex": 1, "explanation": "An observation is directly supported by the data; an inference is an explanation needing more evidence. 'The supplier caused it' interprets the cause and requires proof.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "fu8zq61metnn0" }, { "id": "ot-data-sensitivity", "shape": "mcq", "tags": [ "sensitivity-analysis", "data" ], "prompt": { "modality": "text", "value": "What does a sensitivity analysis do?" }, "options": [ { "modality": "text", "value": "Produces a single most-likely forecast number", "short": "Single most-likely forecast" }, { "modality": "text", "value": "Averages several months together to smooth the trend", "short": "Averages months to smooth trend" }, { "modality": "text", "value": "Tests how a result changes when one important input changes", "short": "How result reacts to one input" }, { "modality": "text", "value": "Removes outliers before any analysis begins", "short": "Removes outliers beforehand" } ], "correctIndex": 2, "explanation": "Sensitivity analysis changes one important assumption at a time to show whether a recommendation is robust or fragile.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "fgbaibffwgnh" }, { "id": "ot-data-scenarios", "shape": "mcq", "tags": [ "scenarios", "forecasting" ], "prompt": { "modality": "text", "value": "What is the main benefit of using base, upside, and downside cases instead of one forecast?" }, "options": [ { "modality": "text", "value": "It guarantees a more accurate prediction", "short": "Guarantees accurate prediction" }, { "modality": "text", "value": "It hides uncertainty from a reviewer", "short": "Hides uncertainty from reviewer" }, { "modality": "text", "value": "It reveals which assumptions drive the decision", "short": "Shows which assumptions matter" }, { "modality": "text", "value": "It removes the need to check data quality", "short": "Removes data-quality checks" } ], "correctIndex": 2, "explanation": "Scenarios expose the assumptions carrying the result and show where a contingency plan is needed; they do not guarantee accuracy or replace data checks.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1cu46ik1tw4dx8" }, { "id": "ot-data-inspect", "shape": "mcq", "tags": [ "data-quality", "data" ], "prompt": { "modality": "text", "value": "Before asking AI for conclusions, which is NOT part of inspecting data quality?" }, "options": [ { "modality": "text", "value": "Identifying missing values", "short": "Identifying missing values" }, { "modality": "text", "value": "Choosing the chart color scheme", "short": "Choosing chart color scheme" }, { "modality": "text", "value": "Clarifying ambiguous column definitions", "short": "Clarifying column definitions" }, { "modality": "text", "value": "Checking for duplicate records", "short": "Checking duplicate records" } ], "correctIndex": 1, "explanation": "Data-quality inspection covers duplicates, missing values, unclear units, and column meanings. A chart's colors are presentation, not data quality.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "15vwcf9135k11f" }, { "id": "ot-marketing-brand-brief", "shape": "mcq", "tags": [ "brand brief", "marketing" ], "prompt": { "modality": "text", "value": "Which of the following is NOT typically part of a one-page brand brief?" }, "options": [ { "modality": "text", "value": "Preferred terminology and phrases to avoid", "short": "Preferred terms & words to avoid" }, { "modality": "text", "value": "A month-by-month sales revenue forecast", "short": "Monthly sales revenue forecast" }, { "modality": "text", "value": "The desired tone of voice", "short": "The desired tone of voice" }, { "modality": "text", "value": "Three proof points that support your promise", "short": "Proof points for your promise" } ], "correctIndex": 1, "explanation": "A brand brief names the customer, promise, tone, proof points, preferred and prohibited terms, and examples. A revenue forecast is a financial projection, not part of the brief.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1jbf14ljxw5ur" }, { "id": "ot-marketing-content-atom", "shape": "mcq", "tags": [ "content atom", "adaptation" ], "prompt": { "modality": "text", "value": "You wrote one detailed customer success story. Following the content-atom approach, what should you do with it?" }, "options": [ { "modality": "text", "value": "Post it only on the one channel where it performs best", "short": "Post on one best channel only" }, { "modality": "text", "value": "Discard it after a single use to keep your content fresh", "short": "Discard after a single use" }, { "modality": "text", "value": "Adapt it into an email, social posts, a sales handout, and video scripts, each tailored to its channel", "short": "Tailor it to each channel" }, { "modality": "text", "value": "Publish the identical text on every channel to stay consistent", "short": "Post identical text everywhere" } ], "correctIndex": 2, "explanation": "A content atom is one substantial source asset adapted into multiple channel-specific pieces. Adaptation matters because each channel serves a different context.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "u7mxqjg3dl11" }, { "id": "ot-marketing-copyright", "shape": "mcq", "tags": [ "copyright", "authorship" ], "prompt": { "modality": "text", "value": "According to the U.S. Copyright Office, when can AI-assisted work qualify for copyright protection?" }, "options": [ { "modality": "text", "value": "Whenever the prompt is detailed enough to guide the output", "short": "When the prompt is detailed" }, { "modality": "text", "value": "Never, because AI was involved at some stage", "short": "Never, since AI was involved" }, { "modality": "text", "value": "When a human contributes sufficient expressive authorship", "short": "Human adds expressive authorship" }, { "modality": "text", "value": "Automatically, because all AI output is protected", "short": "Automatically—AI is protected" } ], "correctIndex": 2, "explanation": "The Copyright Office says human expressive authorship can make AI-assisted work protectable, but prompts alone generally do not establish authorship over a purely generated output.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "lopdwjyuik0p" }, { "id": "ot-marketing-proof-point", "shape": "mcq", "tags": [ "proof point", "claims" ], "prompt": { "modality": "text", "value": "In a brand brief, what is a 'proof point'?" }, "options": [ { "modality": "text", "value": "Specific evidence that supports a marketing claim", "short": "Evidence supporting a claim" }, { "modality": "text", "value": "The overall personality and language of the brand", "short": "The brand's overall voice" }, { "modality": "text", "value": "A channel-specific version of a content atom", "short": "A channel-specific atom" }, { "modality": "text", "value": "A phrase the brand has agreed never to use", "short": "A phrase the brand avoids" } ], "correctIndex": 0, "explanation": "A proof point is concrete evidence backing a claim. The other options describe prohibited terms, brand voice, and an adapted content atom.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1okuvt91frtmpt" }, { "id": "ot-cs-escalation-rule", "shape": "mcq", "tags": [ "escalation", "handoff" ], "prompt": { "modality": "text", "value": "What is an escalation rule in a customer-service workflow?" }, "options": [ { "modality": "text", "value": "A cap on how many messages a customer may send per day", "short": "A daily cap on messages sent" }, { "modality": "text", "value": "A ranking of prospects by how likely they are to buy", "short": "Ranking prospects by buy odds" }, { "modality": "text", "value": "A defined condition that stops automation and transfers a case to human review", "short": "Condition sending case to human" }, { "modality": "text", "value": "A script that lets the chatbot answer harder questions on its own", "short": "Chatbot answers hard questions" } ], "correctIndex": 2, "explanation": "An escalation rule tells the workflow when to stop and hand a case to a person. The last option describes lead qualification, not escalation.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1s7py0abkgx4u" }, { "id": "ot-cs-escalation-trigger", "shape": "mcq", "tags": [ "escalation", "triggers" ], "prompt": { "modality": "text", "value": "Which of these is LEAST likely to be set as an escalation trigger?" }, "options": [ { "modality": "text", "value": "A request involving sensitive personal information", "short": "Request with sensitive data" }, { "modality": "text", "value": "A customer making a legal threat", "short": "A customer's legal threat" }, { "modality": "text", "value": "A routine question about current store hours", "short": "Routine store-hours question" }, { "modality": "text", "value": "A refund request above your defined limit", "short": "Refund above your set limit" } ], "correctIndex": 2, "explanation": "Threats, refunds above a limit, and sensitive information are common triggers to route to a person. A routine hours question is exactly what automation should answer.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1ywsigey2ikv2" }, { "id": "ot-cs-grounding", "shape": "mcq", "tags": [ "grounding", "approved knowledge" ], "prompt": { "modality": "text", "value": "A customer-facing AI assistant should base its answers primarily on what?" }, "options": [ { "modality": "text", "value": "The most popular answers found on competitor websites", "short": "Popular competitor answers" }, { "modality": "text", "value": "Whatever reply sounds the most confident and helpful", "short": "Whatever sounds most confident" }, { "modality": "text", "value": "The business's approved policies, pricing, hours, and canonical answers", "short": "Approved policy, price, hours" }, { "modality": "text", "value": "The general knowledge built into the language model", "short": "The model's general knowledge" } ], "correctIndex": 2, "explanation": "Customer-facing AI must be grounded in approved business records. General model knowledge is not a substitute for your own policies and pricing.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "t5ua2ktclqfm" }, { "id": "ot-cs-personalization", "shape": "mcq", "tags": [ "personalization", "privacy" ], "prompt": { "modality": "text", "value": "Which use of personalization in a customer reply is appropriate rather than 'creepy'?" }, "options": [ { "modality": "text", "value": "Revealing a hidden profile the system quietly built about them", "short": "A hidden profile built on them" }, { "modality": "text", "value": "Referencing the order the customer just placed with you", "short": "The order they just placed" }, { "modality": "text", "value": "Bringing up data the customer never expected you to have", "short": "Unexpected data about them" }, { "modality": "text", "value": "Mentioning an unrelated purchase the customer made elsewhere", "short": "An unrelated purchase elsewhere" } ], "correctIndex": 1, "explanation": "Good personalization uses information the customer knowingly supplied for the relationship. The other options reveal hidden or unrelated data the customer did not expect you to use.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1k3fipki1ychq" }, { "id": "ot-sop-exception-path", "shape": "mcq", "tags": [ "sop", "exceptions" ], "prompt": { "modality": "text", "value": "In a standard operating procedure, what does the exception path describe?" }, "options": [ { "modality": "text", "value": "The normal sequence followed when nothing unusual happens", "short": "Normal sequence, nothing odd" }, { "modality": "text", "value": "Instructions for unusual conditions or failures", "short": "Steps for unusual conditions" }, { "modality": "text", "value": "A predefined trigger and matching automatic action", "short": "A trigger and auto action" }, { "modality": "text", "value": "A required human confirmation before spending money", "short": "Human OK before spending" } ], "correctIndex": 1, "explanation": "The exception path covers unusual conditions or failures (blank field, objection, system down). The normal sequence is the happy path.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "11p2arq1113ay4" }, { "id": "ot-sop-not-complete", "shape": "mcq", "tags": [ "sop", "completeness" ], "prompt": { "modality": "text", "value": "Which of these is NOT something a well-formed operating procedure should include?" }, "options": [ { "modality": "text", "value": "An assigned owner and a review date", "short": "An owner and review date" }, { "modality": "text", "value": "A guarantee that no experienced worker input is ever needed", "short": "No worker input ever needed" }, { "modality": "text", "value": "Exception paths for missing information or failures", "short": "Exception paths for failures" }, { "modality": "text", "value": "Steps, decisions, and safety controls", "short": "Steps, decisions, controls" } ], "correctIndex": 1, "explanation": "The experienced worker, not the model, is the authority, so their input is essential. Good procedures do include steps, exceptions, an owner, and a review date.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "16im1nvnurilt" }, { "id": "ot-sop-hidden-decisions", "shape": "mcq", "tags": [ "sop", "hidden-decisions" ], "prompt": { "modality": "text", "value": "When AI drafts a procedure from a transcript, what does the guide say to do with rules you are unsure about?" }, "options": [ { "modality": "text", "value": "Drop the uncertain step to keep the happy path clean", "short": "Drop the uncertain step" }, { "modality": "text", "value": "Ship the draft and fix problems only after users report them", "short": "Ship now, fix when users report" }, { "modality": "text", "value": "Let the model fill the gap with its best guess so the draft stays complete", "short": "Let AI guess to fill the gap" }, { "modality": "text", "value": "Mark each uncertain rule with a question and resolve it with the process owner", "short": "Flag uncertain rules, ask owner" } ], "correctIndex": 3, "explanation": "The guide says mark every uncertain rule with a question instead of filling the gap, then resolve it with the process owner. Guessing hides the real expertise.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "10p6noh1e4p26d" }, { "id": "ot-sop-test-person", "shape": "mcq", "tags": [ "sop", "testing" ], "prompt": { "modality": "text", "value": "After drafting a procedure, how does the guide recommend checking that it actually works?" }, "options": [ { "modality": "text", "value": "Publish it immediately without review to save time", "short": "Publish now, skip review" }, { "modality": "text", "value": "Replace it with a generic AI answer about the industry", "short": "Swap in a generic AI answer" }, { "modality": "text", "value": "Ask a competitor to validate the wording", "short": "Have a competitor validate" }, { "modality": "text", "value": "Test it with a different person and watch where they pause or guess", "short": "Test it with a fresh person" } ], "correctIndex": 3, "explanation": "Testing with a different person reveals where they pause, guess, or need undocumented access, so you can revise the procedure before relying on it.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1ad7ga3phxf3f" }, { "id": "ot-auto-agent-vs-rule", "shape": "mcq", "tags": [ "agents", "automation" ], "prompt": { "modality": "text", "value": "What distinguishes an AI agent from rule-based automation?" }, "options": [ { "modality": "text", "value": "It never requires any permissions or boundaries", "short": "Needs no permissions or limits" }, { "modality": "text", "value": "It only copies form submissions into a CRM", "short": "Only copies forms into a CRM" }, { "modality": "text", "value": "It can choose steps and use tools toward a goal", "short": "Picks steps and tools for a goal" }, { "modality": "text", "value": "It follows one predefined trigger and one fixed action", "short": "One trigger, one fixed action" } ], "correctIndex": 2, "explanation": "An agent selects steps and uses tools toward a goal, while rule-based automation runs a predefined trigger and action. That flexibility is why agents need firmer boundaries.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1ck45eg1ppfj02" }, { "id": "ot-auto-approval-gate", "shape": "mcq", "tags": [ "approval-gate", "controls" ], "prompt": { "modality": "text", "value": "What is an approval gate in an automated workflow?" }, "options": [ { "modality": "text", "value": "A rule that triggers an action automatically with no review", "short": "Auto trigger, no review" }, { "modality": "text", "value": "A required human confirmation before a consequential action", "short": "Human sign-off before action" }, { "modality": "text", "value": "A log recording inputs, outputs, actions, and errors", "short": "A log of inputs and actions" }, { "modality": "text", "value": "The least amount of access a workflow is granted", "short": "The least access granted" } ], "correctIndex": 1, "explanation": "An approval gate is a required human confirmation before consequential actions like payments or external messages. Logs and least privilege are separate controls.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "kdyhsj2vkazn" }, { "id": "ot-auto-least-privilege", "shape": "mcq", "tags": [ "permissions", "least-privilege" ], "prompt": { "modality": "text", "value": "Following least privilege, which permissions should a beginner automation usually start with?" }, "options": [ { "modality": "text", "value": "Read and draft permissions", "short": "Read and draft permissions" }, { "modality": "text", "value": "Full send and spend permissions", "short": "Send and spend permissions" }, { "modality": "text", "value": "Unlimited access to every connected system", "short": "Unlimited system permissions" }, { "modality": "text", "value": "Delete and account-change permissions", "short": "Delete and account permissions" } ], "correctIndex": 0, "explanation": "Least privilege means beginning with read and draft permissions and expanding only when measured performance justifies it. Send, spend, and delete are the riskiest.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1myhmvi1rxcew2" }, { "id": "ot-auto-rule-enough", "shape": "mcq", "tags": [ "automation", "rule-vs-agent" ], "prompt": { "modality": "text", "value": "For which task is simple rule-based automation usually enough, with no AI agent needed?" }, "options": [ { "modality": "text", "value": "Summarizing a long meeting transcript", "short": "Summarize a meeting transcript" }, { "modality": "text", "value": "Drafting a tailored reply to unusual feedback", "short": "Draft a reply to odd feedback" }, { "modality": "text", "value": "Copying new form submissions into a CRM", "short": "Copy form submissions to CRM" }, { "modality": "text", "value": "Classifying an ambiguous customer exception", "short": "Classify an ambiguous case" } ], "correctIndex": 2, "explanation": "Rule-based automation reliably handles fixed triggers like copying form submissions. AI earns its place only when a workflow must interpret variable text, classify exceptions, summarize, or draft.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1mujfqpaub22j" }, { "id": "ot-high-stakes-material-term", "shape": "mcq", "tags": [ "contracts", "definitions" ], "prompt": { "modality": "text", "value": "In a contract, what best describes a 'material term'?" }, "options": [ { "modality": "text", "value": "Any clause written in legal boilerplate language", "short": "Any boilerplate legal clause" }, { "modality": "text", "value": "The physical materials a signed paper contract is printed on", "short": "The paper it's printed on" }, { "modality": "text", "value": "A provision important enough to affect cost, rights, obligations, risk, or the decision to sign", "short": "Provision affecting cost/risk" }, { "modality": "text", "value": "A minor formatting detail that has no bearing on the agreement", "short": "A minor formatting detail" } ], "correctIndex": 2, "explanation": "A material term is significant enough to influence cost, rights, obligations, risk, or whether to sign. The others describe boilerplate, paper stock, or trivial details, which are not material.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1m0qlxs1ve948m" }, { "id": "ot-high-stakes-reconcile", "shape": "mcq", "tags": [ "finance", "reconciliation" ], "prompt": { "modality": "text", "value": "Which statement about using AI for financial work is INCORRECT?" }, "options": [ { "modality": "text", "value": "AI can help surface anomalies and organize a review", "short": "AI can help surface anomalies" }, { "modality": "text", "value": "A plausible-looking dashboard total proves the accounts actually agree", "short": "Dashboard total = accounts agree" }, { "modality": "text", "value": "Summaries should be tied back to statements, invoices, payroll, or the ledger", "short": "Tie summaries back to records" }, { "modality": "text", "value": "A bookkeeper or CPA should resolve flagged discrepancies", "short": "A CPA resolves discrepancies" } ], "correctIndex": 1, "explanation": "A plausible total is not a reconciled total; an attractive dashboard is not proof records agree. The other three reflect the guide's reconcile-to-source approach.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "ftvs3lroj1z" }, { "id": "ot-high-stakes-tax", "shape": "mcq", "tags": [ "tax", "limitations" ], "prompt": { "modality": "text", "value": "Why should a small business avoid relying on general AI for complex tax answers?" }, "options": [ { "modality": "text", "value": "AI cannot read any tax forms or instructions", "short": "AI can't read tax forms" }, { "modality": "text", "value": "Tax questions are too simple to need any preparation", "short": "Tax questions are too simple" }, { "modality": "text", "value": "Current law, jurisdiction, structure, elections, and records all matter", "short": "Law, structure, records matter" }, { "modality": "text", "value": "Tax software is always cheaper than any AI tool", "short": "Tax software is always cheaper" } ], "correctIndex": 2, "explanation": "General AI may omit elections, exceptions, jurisdictional rules, or recent changes; current law, structure, and records matter. Use AI to prepare records and questions, not to manufacture certainty.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1obxpy81ql934a" }, { "id": "ot-high-stakes-responsibility", "shape": "mcq", "tags": [ "workflow", "responsibility" ], "prompt": { "modality": "text", "value": "Which task is appropriate for AI's first pass rather than a qualified human's final decision?" }, "options": [ { "modality": "text", "value": "Summarizing a lease and drafting questions for counsel", "short": "Summarize a lease for counsel" }, { "modality": "text", "value": "Authorizing a payment based on a forecast", "short": "Authorize a forecast payment" }, { "modality": "text", "value": "Accepting responsibility for a tax filing", "short": "Accept a tax filing's outcome" }, { "modality": "text", "value": "Approving a loan commitment on the owner's behalf", "short": "Approve a loan commitment" } ], "correctIndex": 0, "explanation": "AI's first pass is to summarize, classify, compare, or prepare questions. The final filing, commitment, payment, or approval stays with a qualified, responsible person.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "jub41h1euxiod" }, { "id": "ot-people-adverse-impact", "shape": "mcq", "tags": [ "hiring", "definitions" ], "prompt": { "modality": "text", "value": "What is 'adverse impact' in employment?" }, "options": [ { "modality": "text", "value": "A substantially different negative outcome for a protected group from a seemingly neutral practice", "short": "Neutral rule hurts a group" }, { "modality": "text", "value": "A deliberate, openly stated policy of excluding a group", "short": "Openly excluding a group" }, { "modality": "text", "value": "Any negative feedback an employee receives during a review", "short": "Negative feedback in a review" }, { "modality": "text", "value": "A drop in overall company hiring during a slow year", "short": "A drop in overall hiring" } ], "correctIndex": 0, "explanation": "Adverse impact is a substantially different negative outcome for a protected group caused by a seemingly neutral practice. It need not be intentional, which distinguishes it from an openly discriminatory policy.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "sl9pd8ig9k4k" }, { "id": "ot-people-eeoc", "shape": "mcq", "tags": [ "compliance", "accommodation" ], "prompt": { "modality": "text", "value": "The EEOC and Department of Justice warn that algorithmic hiring tools may do what?" }, "options": [ { "modality": "text", "value": "Guarantee fully bias-free hiring decisions", "short": "Guarantee bias-free hiring" }, { "modality": "text", "value": "Screen out people with disabilities", "short": "Screen out disabled applicants" }, { "modality": "text", "value": "Eliminate the need for any human reviewer", "short": "Remove the need for human review" }, { "modality": "text", "value": "Automatically satisfy all employment law", "short": "Auto-satisfy employment law" } ], "correctIndex": 1, "explanation": "The EEOC and DOJ warn such tools may screen out people with disabilities, so employers should offer an accommodation process and an alternative. The other options overstate what the tools achieve.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1mpxg5y1xit9ek" }, { "id": "ot-people-human-review", "shape": "mcq", "tags": [ "human-review", "hiring" ], "prompt": { "modality": "text", "value": "What does 'meaningful human review' of an AI hiring recommendation require?" }, "options": [ { "modality": "text", "value": "Trusting the tool's output because it looks objective", "short": "Trusting the tool's output" }, { "modality": "text", "value": "Reviewing only the candidates the tool already advanced", "short": "Review only advanced candidates" }, { "modality": "text", "value": "Authority, time, relevant information, and ability to challenge the recommendation", "short": "Authority, info, time to object" }, { "modality": "text", "value": "Clicking approve quickly to keep the process moving", "short": "Clicking approve quickly" } ], "correctIndex": 2, "explanation": "Meaningful review is more than clicking approve; the reviewer needs authority, time, relevant information, and the ability to challenge the result, then document the criteria and decision.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "pl41ev1pdc5pz" }, { "id": "ot-people-protected", "shape": "mcq", "tags": [ "protected-traits", "hiring" ], "prompt": { "modality": "text", "value": "Which task should you NOT ask an AI hiring tool to do?" }, "options": [ { "modality": "text", "value": "Organize interview notes for a human reviewer", "short": "Organize interview notes" }, { "modality": "text", "value": "Remove jargon from a job posting", "short": "Remove jargon from a posting" }, { "modality": "text", "value": "Infer a candidate's health, age, or family status", "short": "Infer health, age, or family" }, { "modality": "text", "value": "Map applications to documented, job-related requirements", "short": "Map apps to job requirements" } ], "correctIndex": 2, "explanation": "Do not ask AI to infer personality, health, family status, age, race, religion, or other protected traits. The other tasks are administrative uses tied to documented, job-related criteria.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "18f0mhwkobo7i" }, { "id": "ot-data-protection-minimization", "shape": "mcq", "tags": [ "data-minimization", "privacy" ], "prompt": { "modality": "text", "value": "What does practicing data minimization mean when using an AI service?" }, "options": [ { "modality": "text", "value": "Share only the fields actually needed for the task", "short": "Share only needed fields" }, { "modality": "text", "value": "Keep every record indefinitely in case it is useful later", "short": "Keep every record forever" }, { "modality": "text", "value": "Encrypt data before pasting it into any tool", "short": "Encrypt data before pasting" }, { "modality": "text", "value": "Share all available data so the AI has full context", "short": "Share all available data" } ], "correctIndex": 0, "explanation": "Data minimization means sharing only the fields needed for the task, which reduces exposure and often improves focus. Sharing everything raises risk.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1wajmu11smfwj7" }, { "id": "ot-data-protection-plan-controls", "shape": "mcq", "tags": [ "plan-controls", "vendor" ], "prompt": { "modality": "text", "value": "Your team upgrades from a free to a business AI plan. What should you assume about its training and retention settings?" }, "options": [ { "modality": "text", "value": "They may differ, so verify the current terms for that exact product", "short": "They may differ; verify terms" }, { "modality": "text", "value": "Free plans always offer stronger privacy", "short": "Free plans are more private" }, { "modality": "text", "value": "They are identical across all plan tiers", "short": "Identical across all tiers" }, { "modality": "text", "value": "Business plans never retain any data", "short": "Business plans never retain data" } ], "correctIndex": 0, "explanation": "Training, retention, and sharing can differ by product and account type, so you must verify the current controls for the exact product. Plan labels alone are not data controls.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "ykzadb1ut3vuj" }, { "id": "ot-data-protection-public", "shape": "mcq", "tags": [ "classification", "public-data" ], "prompt": { "modality": "text", "value": "Using the labels public, internal, confidential, and restricted, which item is public data?" }, "options": [ { "modality": "text", "value": "A customer contact list", "short": "Customer contact list" }, { "modality": "text", "value": "An employee payroll file", "short": "Employee payroll file" }, { "modality": "text", "value": "A signed contract under confidentiality", "short": "Signed confidential contract" }, { "modality": "text", "value": "A marketing brochure published on your website", "short": "Marketing brochure on website" } ], "correctIndex": 3, "explanation": "Public data is information intentionally available to anyone, like published marketing copy. Customer lists, payroll files, and confidential contracts are restricted business information.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "8bq3dtputd93" }, { "id": "ot-data-protection-retention", "shape": "mcq", "tags": [ "retention", "definition" ], "prompt": { "modality": "text", "value": "In vendor and data terms, what does 'data retention' refer to?" }, "options": [ { "modality": "text", "value": "How long a service or business keeps information before deletion or archival", "short": "How long info is stored" }, { "modality": "text", "value": "The process of encrypting stored records", "short": "Encrypting stored records" }, { "modality": "text", "value": "The number of people who can access a dataset", "short": "How many can access data" }, { "modality": "text", "value": "Whether inputs are used to train a model", "short": "Whether inputs train a model" } ], "correctIndex": 0, "explanation": "Data retention is how long information is kept before deletion or archival. Access counts and training use are separate vendor questions.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "18ixyvv11fw1q9" }, { "id": "ot-governance-consequential", "shape": "mcq", "tags": [ "consequential-action", "risk-tiers" ], "prompt": { "modality": "text", "value": "In AI governance, which best describes a consequential action?" }, "options": [ { "modality": "text", "value": "An action that can materially affect money, rights, safety, or trust", "short": "Affects money, rights, safety" }, { "modality": "text", "value": "An action that can be corrected later with limited harm or cost", "short": "Correctable with limited harm" }, { "modality": "text", "value": "Any step a person happens to review before it runs", "short": "A step someone reviews first" }, { "modality": "text", "value": "A task that stays entirely inside internal brainstorming", "short": "Stays in internal brainstorm" } ], "correctIndex": 0, "explanation": "A consequential action can materially affect money, rights, safety, or trust, so it needs stronger controls. An action correctable with limited harm is a reversible one, not a consequential one.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "hzh9i3f5ij31" }, { "id": "ot-governance-human-loop", "shape": "mcq", "tags": [ "human-in-loop", "oversight" ], "prompt": { "modality": "text", "value": "Where should a human in the loop be placed to be effective?" }, "options": [ { "modality": "text", "value": "At a point where review can still change the outcome", "short": "Where review changes outcome" }, { "modality": "text", "value": "After the action is complete and irreversible", "short": "After action is irreversible" }, { "modality": "text", "value": "Only during the initial software setup", "short": "Only during initial setup" }, { "modality": "text", "value": "Wherever it slows the workflow the least", "short": "Where it slows workflow least" } ], "correctIndex": 0, "explanation": "A human in the loop must sit where review can still change the outcome, with approval gates before consequential actions. Review after an irreversible action cannot prevent harm.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "qzdplk1qx9pc2" }, { "id": "ot-governance-incident", "shape": "mcq", "tags": [ "incident", "response" ], "prompt": { "modality": "text", "value": "Which action is NOT part of a sound AI incident plan?" }, "options": [ { "modality": "text", "value": "Notifying the responsible people", "short": "Notify the responsible people" }, { "modality": "text", "value": "Correcting affected records or messages", "short": "Correct affected records" }, { "modality": "text", "value": "Stopping the workflow", "short": "Stop the running workflow" }, { "modality": "text", "value": "Deleting all logs to reduce liability", "short": "Delete all logs for liability" } ], "correctIndex": 3, "explanation": "An incident plan says to preserve records, not delete them. Stopping the workflow, correcting affected data, and notifying owners are all core steps.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "85prjjylwb4v" }, { "id": "ot-governance-risk-tier", "shape": "mcq", "tags": [ "risk-tiers", "governance" ], "prompt": { "modality": "text", "value": "A cafe uses AI to brainstorm in-house promotion ideas that never leave the shop and are easy to redo. Which risk tier fits best?" }, "options": [ { "modality": "text", "value": "Low risk" }, { "modality": "text", "value": "Prohibited use" }, { "modality": "text", "value": "Medium risk" }, { "modality": "text", "value": "High risk" } ], "correctIndex": 0, "explanation": "Internal, low-stakes, easily undone work is low risk. Medium reaches customers, high affects money or rights, and prohibited uses are those whose risk cannot be managed.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "n5g3lq11i1q0g" }, { "id": "ot-pilot-define", "shape": "mcq", "tags": [ "pilot", "scope" ], "prompt": { "modality": "text", "value": "What best describes a 30-day pilot?" }, "options": [ { "modality": "text", "value": "A permanent rollout across every team at once", "short": "Permanent rollout to all teams" }, { "modality": "text", "value": "A demo meant to show that AI is exciting", "short": "A demo to show AI is exciting" }, { "modality": "text", "value": "An open-ended trial with no fixed end date", "short": "Open-ended trial, no end date" }, { "modality": "text", "value": "A limited implementation with a defined owner, scope, period, controls, and success criteria", "short": "Limited, owner, scope, criteria" } ], "correctIndex": 3, "explanation": "A pilot is a limited, controlled implementation that exists to create evidence, not to prove AI is exciting or to launch everywhere immediately.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "5kzfff8yq0xz" }, { "id": "ot-pilot-baseline", "shape": "mcq", "tags": [ "baseline", "measurement" ], "prompt": { "modality": "text", "value": "Which measurement is NOT part of recording a baseline before the pilot?" }, "options": [ { "modality": "text", "value": "Error rate and turnaround", "short": "Error rate and turnaround" }, { "modality": "text", "value": "Current time and cost", "short": "Current time and cost" }, { "modality": "text", "value": "Backlog or customer outcome", "short": "Backlog or customer outcome" }, { "modality": "text", "value": "The final expansion decision", "short": "Final expansion decision" } ], "correctIndex": 3, "explanation": "A baseline captures current time, cost, error rate, turnaround, backlog, and customer outcome. The expansion decision comes at the end, after comparing results to that baseline.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1sl46mo1qxnmsi" }, { "id": "ot-pilot-parallel", "shape": "mcq", "tags": [ "parallel-run", "testing" ], "prompt": { "modality": "text", "value": "In the second week, how should the AI workflow be tested against the current process?" }, "options": [ { "modality": "text", "value": "By replacing the current process entirely", "short": "Replace the current process" }, { "modality": "text", "value": "By running a parallel test beside the current process", "short": "Run parallel test beside it" }, { "modality": "text", "value": "By expanding scope before any review", "short": "Expand scope before review" }, { "modality": "text", "value": "By skipping the current process to save time", "short": "Skip current process for speed" } ], "correctIndex": 1, "explanation": "A parallel run works the AI beside the existing process so you can record where it succeeds, fails, and how long review takes without risking live work.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "4hyxtl1kma52j" }, { "id": "ot-pilot-inventory", "shape": "mcq", "tags": [ "task-inventory", "selection" ], "prompt": { "modality": "text", "value": "When building the first-week task inventory, which workflow should you choose for the pilot?" }, "options": [ { "modality": "text", "value": "A frequent, structured, reviewable, low-risk workflow", "short": "Frequent, structured, low-risk" }, { "modality": "text", "value": "A rare, one-off task with high stakes", "short": "Rare, one-off, high-stakes task" }, { "modality": "text", "value": "Whichever task competitors are automating", "short": "Whatever competitors automate" }, { "modality": "text", "value": "The most creative and unpredictable task available", "short": "Creative, unpredictable task" } ], "correctIndex": 0, "explanation": "The pilot should target a frequent, structured, reviewable, low-risk workflow so results are measurable and mistakes are recoverable.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1x66pax1h4zrkb" }, { "id": "ot-os-policy", "shape": "mcq", "tags": [ "ai-policy", "governance" ], "prompt": { "modality": "text", "value": "Which item does NOT belong in a short one-page AI use policy?" }, "options": [ { "modality": "text", "value": "A guaranteed monthly revenue target", "short": "Guaranteed monthly revenue goal" }, { "modality": "text", "value": "External-use rules and an owner", "short": "External-use rules and owner" }, { "modality": "text", "value": "Review rules and incident reporting", "short": "Review rules & incident reports" }, { "modality": "text", "value": "Approved tools and prohibited data", "short": "Approved tools, prohibited data" } ], "correctIndex": 0, "explanation": "A beginner AI policy covers approved tools, prohibited data, review rules, external use, incidents, and responsibility. A revenue target is not part of a use policy.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1g5nt921cvev2w" }, { "id": "ot-os-killswitch", "shape": "mcq", "tags": [ "kill-switch", "controls" ], "prompt": { "modality": "text", "value": "What is the purpose of a kill switch in an AI operating system?" }, "options": [ { "modality": "text", "value": "To require human approval before actions", "short": "Require approval before actions" }, { "modality": "text", "value": "To store tested prompts and examples", "short": "Store tested prompts/examples" }, { "modality": "text", "value": "To rapidly shut the workflow down", "short": "Rapidly shut the workflow down" }, { "modality": "text", "value": "To log every action for accountability", "short": "Log every action for record" } ], "correctIndex": 2, "explanation": "A kill switch exists for rapid shutdown. The audit log handles accountability, the prompt library stores prompts, and the approval gate handles human sign-off.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "v07y6uhdnytw" }, { "id": "ot-os-escalation", "shape": "mcq", "tags": [ "escalation-rule", "exceptions" ], "prompt": { "modality": "text", "value": "An escalation rule tells the workflow what to do in which situation?" }, "options": [ { "modality": "text", "value": "When outputs are routine and low-risk", "short": "Routine, low-risk outputs" }, { "modality": "text", "value": "When the subscription is due for renewal", "short": "Subscription is due to renew" }, { "modality": "text", "value": "When a prompt needs to be rewritten", "short": "A prompt needs rewriting" }, { "modality": "text", "value": "When uncertainty or an exception arises", "short": "Uncertainty or exception" } ], "correctIndex": 3, "explanation": "An escalation rule routes uncertainty and exceptions to a person, rather than letting the AI proceed unchecked on cases it cannot handle.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "60ei1u1kngawy" }, { "id": "ot-os-auditlog", "shape": "mcq", "tags": [ "audit-log", "accountability" ], "prompt": { "modality": "text", "value": "Which control in the operating system exists primarily for accountability?" }, "options": [ { "modality": "text", "value": "The approval gate" }, { "modality": "text", "value": "The prompt library" }, { "modality": "text", "value": "The audit log" }, { "modality": "text", "value": "The kill switch" } ], "correctIndex": 2, "explanation": "An audit log records what happened for accountability. The approval gate blocks consequential actions, the library stores prompts, and the kill switch enables shutdown.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "i09byg1swaup8" }, { "id": "ot-cov-capabilities-1", "shape": "mcq", "tags": [ "verification", "evidence", "ai-limits" ], "prompt": { "modality": "text", "value": "Two surveys report very different AI-adoption rates for small businesses. What should you check before citing either number?" }, "options": [ { "modality": "text", "value": "Whether the source sounds confident" }, { "modality": "text", "value": "The population surveyed and how 'use' was defined" }, { "modality": "text", "value": "Which number is higher" }, { "modality": "text", "value": "The length of the article" } ], "correctIndex": 1, "explanation": "Adoption figures vary by population and definition, so a material statistic needs its source and terms checked before it is repeated.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "hbv35f1kmcidl" }, { "id": "ot-cov-capabilities-2", "shape": "mcq", "tags": [ "verification", "evidence", "ai-limits" ], "prompt": { "modality": "text", "value": "An AI-generated report states a precise revenue increase with no traceable source. What should you do before acting on it?" }, "options": [ { "modality": "text", "value": "Round the number down slightly" }, { "modality": "text", "value": "Trust it because the number is precise" }, { "modality": "text", "value": "Verify it against an independent source first" }, { "modality": "text", "value": "Ask a colleague if it sounds plausible" } ], "correctIndex": 2, "explanation": "A precise-sounding figure without a traceable source is a warning sign; material claims must be independently verified before use.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "weecx2p3iw3a" }, { "id": "ot-cov-iteration-1", "shape": "mcq", "tags": [ "prompt-library", "iteration", "reuse" ], "prompt": { "modality": "text", "value": "A prompt finally produces reliable customer emails. What should you save so the workflow is reusable?" }, "options": [ { "modality": "text", "value": "A screenshot of the AI conversation" }, { "modality": "text", "value": "Instructions, sources, examples, and checklist together" }, { "modality": "text", "value": "Nothing; regenerate from scratch each time" }, { "modality": "text", "value": "Only the final email you liked" } ], "correctIndex": 1, "explanation": "A reusable prompt-library asset saves the full package -- instructions, sources, examples, and checklist -- not just one finished output.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1svsjlqs9f7b6" }, { "id": "ot-cov-iteration-2", "shape": "mcq", "tags": [ "prompt-library", "iteration", "reuse" ], "prompt": { "modality": "text", "value": "Which practice keeps a saved prompt-library entry useful as models and policies change?" }, "options": [ { "modality": "text", "value": "Deleting the checklist after the first success" }, { "modality": "text", "value": "Sharing prompts without an owner listed" }, { "modality": "text", "value": "Dating entries and retesting when policy or model changes" }, { "modality": "text", "value": "Copying popular prompts from social media" } ], "correctIndex": 2, "explanation": "Dating entries and retesting them when a model or policy changes keeps a prompt-library asset trustworthy over time.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "cjx1j718m7cgv" }, { "id": "ot-cov-research-1", "shape": "mcq", "tags": [ "triangulation", "research", "evidence" ], "prompt": { "modality": "text", "value": "You need to confirm a local demand claim before relying on it. Which approach best reflects triangulation?" }, "options": [ { "modality": "text", "value": "Averaging three AI-generated guesses" }, { "modality": "text", "value": "Reading the same blog post twice" }, { "modality": "text", "value": "Asking the AI to restate the number more confidently" }, { "modality": "text", "value": "Cross-check sales data, interviews, and a government survey" } ], "correctIndex": 3, "explanation": "Triangulation means checking a claim against multiple independent sources or methods, such as sales data, interviews, and public statistics.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "fup6od10eulbr" }, { "id": "ot-cov-research-2", "shape": "mcq", "tags": [ "triangulation", "research", "evidence" ], "prompt": { "modality": "text", "value": "Why is agreement among independent research methods more convincing than several articles repeating one claim?" }, "options": [ { "modality": "text", "value": "Repeated articles often share one unverified source" }, { "modality": "text", "value": "It takes less time to find" }, { "modality": "text", "value": "More articles always mean more accuracy" }, { "modality": "text", "value": "Independent methods can each fail in the same way" } ], "correctIndex": 0, "explanation": "Articles that repeat one another usually trace back to a single unverified source, so their agreement is not independent confirmation.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1d1jqrp9lmnhv" }, { "id": "ot-cov-customer-service-1", "shape": "mcq", "tags": [ "approved-knowledge", "customer-service", "policy" ], "prompt": { "modality": "text", "value": "A customer assistant is being set up to answer product and policy questions. What should its answers be grounded in?" }, "options": [ { "modality": "text", "value": "Approved policies, product details, hours, and pricing rules" }, { "modality": "text", "value": "The AI's general training knowledge" }, { "modality": "text", "value": "Whatever sounds most helpful" }, { "modality": "text", "value": "Competitor websites" } ], "correctIndex": 0, "explanation": "Customer-facing answers should come from the business's own approved policies and records, not general model knowledge.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "xlo2bt1uwnv0t" }, { "id": "ot-cov-customer-service-2", "shape": "mcq", "tags": [ "approved-knowledge", "customer-service", "policy" ], "prompt": { "modality": "text", "value": "Which source should a customer-facing AI treat as authoritative when answering a pricing question?" }, "options": [ { "modality": "text", "value": "The customer's own estimate" }, { "modality": "text", "value": "A general model's memorized knowledge" }, { "modality": "text", "value": "A confident guess based on similar businesses" }, { "modality": "text", "value": "The business's own approved pricing rules and records" } ], "correctIndex": 3, "explanation": "An approved-knowledge design answers from the business's own current pricing rules and records, not from general or guessed information.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "ftdaxj52ag1f" }, { "id": "ot-cov-people-1", "shape": "mcq", "tags": [ "adverse-impact", "hiring", "fairness" ], "prompt": { "modality": "text", "value": "Why should an owner periodically compare who advances versus who is screened out by a hiring tool?" }, "options": [ { "modality": "text", "value": "To satisfy the vendor's warranty terms" }, { "modality": "text", "value": "To speed up the review process" }, { "modality": "text", "value": "To detect a disparate negative outcome for a protected group" }, { "modality": "text", "value": "To reduce recruiting costs" } ], "correctIndex": 2, "explanation": "A neutral-looking process can still produce adverse impact, so outcomes must be checked for a substantially different negative effect on a protected group.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "134s20hzyk5td" }, { "id": "ot-cov-people-2", "shape": "mcq", "tags": [ "adverse-impact", "hiring", "fairness" ], "prompt": { "modality": "text", "value": "The EEOC has warned that algorithmic hiring tools carry which specific risk?" }, "options": [ { "modality": "text", "value": "Higher advertising costs" }, { "modality": "text", "value": "Reduced applicant volume" }, { "modality": "text", "value": "Screening out qualified people with disabilities" }, { "modality": "text", "value": "Slower processing times" } ], "correctIndex": 2, "explanation": "The EEOC and DOJ have warned that algorithmic tools may screen out people with disabilities, a form of adverse impact requiring an accommodation process.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "naen4jgl1wu1" }, { "id": "ot-cov-data-protection-1", "shape": "mcq", "tags": [ "data-privacy", "data-minimization" ], "prompt": { "modality": "text", "value": "What does data minimization mean when sharing information with an AI tool?" }, "options": [ { "modality": "text", "value": "Convert all data into images before sharing" }, { "modality": "text", "value": "Encrypt every file before upload" }, { "modality": "text", "value": "Delete the AI vendor's account after use" }, { "modality": "text", "value": "Share only the fields needed for the task" } ], "correctIndex": 3, "explanation": "Data minimization means sharing only what a task requires, which reduces exposure of unnecessary information.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "z34giz1p237wf" }, { "id": "ot-cov-data-protection-2", "shape": "mcq", "tags": [ "data-privacy", "data-minimization" ], "prompt": { "modality": "text", "value": "You only need a product category and date for an AI task. What should you do before sharing customer data?" }, "options": [ { "modality": "text", "value": "Ask the vendor to retain the data forever" }, { "modality": "text", "value": "Remove fields the task does not need" }, { "modality": "text", "value": "Share names but keep every other identifier" }, { "modality": "text", "value": "Upload the entire customer file for context" } ], "correctIndex": 1, "explanation": "Minimization means removing fields beyond what the task requires, not just swapping out direct identifiers.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "4awuqqy4tp26" }, { "id": "ot-cov-data-protection-3", "shape": "mcq", "tags": [ "vendor-controls", "data-privacy" ], "prompt": { "modality": "text", "value": "Before using an AI vendor for sensitive work, what should you check?" }, "options": [ { "modality": "text", "value": "Default training use and data retention period" }, { "modality": "text", "value": "Whether the vendor offers phone support" }, { "modality": "text", "value": "How many five-star reviews the tool has" }, { "modality": "text", "value": "Whether the tool offers a free trial period" } ], "correctIndex": 0, "explanation": "Vendor evaluation should cover default training use and data retention period before sensitive data is shared.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "ckzegw1n8e4ek" }, { "id": "ot-cov-data-protection-4", "shape": "mcq", "tags": [ "vendor-controls", "data-privacy" ], "prompt": { "modality": "text", "value": "Per the lesson, which OpenAI product tiers are stated to skip customer inputs and outputs for model training by default?" }, "options": [ { "modality": "text", "value": "Educational classroom accounts" }, { "modality": "text", "value": "Mobile app only" }, { "modality": "text", "value": "Business, Enterprise, and API" }, { "modality": "text", "value": "Free personal accounts" } ], "correctIndex": 2, "explanation": "The guide states OpenAI's Business, Enterprise, and API offerings do not use inputs and outputs for training by default.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1qf3hn01bktnba" }, { "id": "ot-cov-pilot-1", "shape": "mcq", "tags": [ "pilot", "acceptance-criteria" ], "prompt": { "modality": "text", "value": "In a 30-day AI pilot, when should acceptance criteria be defined?" }, "options": [ { "modality": "text", "value": "After results are already compared to competitors" }, { "modality": "text", "value": "During the first week, before running the pilot" }, { "modality": "text", "value": "Only after the pilot ends" }, { "modality": "text", "value": "During the annual budget review" } ], "correctIndex": 1, "explanation": "The guide places acceptance criteria in the first week, alongside the baseline and the data the pilot may use.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1a33ak6p15lgm" }, { "id": "ot-cov-pilot-2", "shape": "mcq", "tags": [ "pilot", "acceptance-criteria" ], "prompt": { "modality": "text", "value": "What role do acceptance criteria play in a 30-day AI pilot?" }, "options": [ { "modality": "text", "value": "Scheduling the parallel run's start date" }, { "modality": "text", "value": "Tracking how much the subscription costs" }, { "modality": "text", "value": "Deciding in advance what results count as success" }, { "modality": "text", "value": "Choosing which employee owns the workflow" } ], "correctIndex": 2, "explanation": "Acceptance criteria set the standard results must meet, so the later expansion decision is evidence-based, not a guess.", "source": { "label": "AI for Small Business — Objective Test" }, "uid": "1xrj4oz101wj3d" } ], "lessons": [ { "id": "aisb-l1", "title": "AI Foundations for an Owner", "order": 0, "studyGuidePath": "/packs/ai-for-small-business/guides/l1-foundations.md", "sources": [ { "label": "U.S. Small Business Administration — AI for small business", "url": "https://www.sba.gov/business-guide/manage-your-business/ai-small-business" }, { "label": "U.S. Census Bureau — AI use by U.S. businesses (May 2026)", "url": "https://www.census.gov/library/stories/2026/05/ai-use-businesses.html" }, { "label": "U.S. Chamber of Commerce — 2025 Empowering Small Business report", "url": "https://www.uschamber.com/technology/artificial-intelligence/u-s-chambers-latest-empowering-small-business-report-shows-majority-of-businesses-in-all-50-states-are-embracing-ai" }, { "label": "Ethan Mollick — Using AI Right Now: A Quick Guide", "url": "https://www.oneusefulthing.org/p/using-ai-right-now-a-quick-guide" } ], "parts": [ { "id": "aisb-l1p1-capabilities", "title": "What AI Can and Cannot Do", "order": 0, "blurb": "A useful mental model without the hype.", "studyGuideAnchor": "capabilities", "itemIds": [ "aisb-fact-pattern-machine", "aisb-fact-probabilistic", "aisb-fact-local-expertise", "aisb-pair-generation", "aisb-pair-analysis", "aisb-def-hallucination", "aisb-mcq-fluency", "aisb-proc-first-experiment", "czr-ai-for-small-business-aisb-def-hallucination", "tf-d-ai-for-small-business-aisb-def-hallucination", "ot-capabilities-hallucination", "ot-capabilities-analysis", "ot-capabilities-fluency-not", "ot-capabilities-stats", "ot-cov-capabilities-1", "ot-cov-capabilities-2" ] }, { "id": "aisb-l1p2-workflow-first", "title": "Start with a Workflow, Not a Tool", "order": 1, "blurb": "Find repetitive work where assistance creates measurable value.", "studyGuideAnchor": "workflow-first", "itemIds": [ "aisb-fact-workflow-first", "aisb-fact-best-first-task", "aisb-fact-augment-before-automate", "aisb-pair-high-fit", "aisb-pair-low-fit", "aisb-def-workflow", "aisb-mcq-first-workflow", "aisb-proc-task-inventory", "czr-ai-for-small-business-aisb-def-workflow", "tf-d-ai-for-small-business-aisb-def-workflow", "ot-workflow-definition", "ot-workflow-worst", "ot-workflow-augment", "ot-workflow-best-task" ] } ], "objectives": [ { "id": "aisb-obj-capabilities", "statement": "Distinguish productive AI assistance from fluent but unsupported output, and verify material claims before acting.", "demonstrationIds": [ "aisb-demo-capabilities" ] }, { "id": "aisb-obj-workflow-fit", "statement": "Choose a frequent, low-risk workflow for a first AI pilot instead of beginning with a tool or high-stakes decision.", "demonstrationIds": [ "aisb-demo-workflow-fit" ] } ], "demonstrations": [ { "id": "aisb-demo-capabilities", "label": "Capabilities, limitations, and hallucinations", "itemIds": [ "aisb-pair-generation", "aisb-pair-analysis", "aisb-mcq-fluency", "czr-ai-for-small-business-aisb-def-hallucination", "ot-cov-capabilities-1", "ot-cov-capabilities-2" ], "requiredCorrect": 3 }, { "id": "aisb-demo-workflow-fit", "label": "Selecting a safe first workflow", "itemIds": [ "aisb-pair-high-fit", "aisb-pair-low-fit", "aisb-mcq-first-workflow", "czr-ai-for-small-business-aisb-def-workflow" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-capabilities", "objectiveId": "aisb-obj-capabilities", "title": "AI Capabilities & Verifying Claims", "mcqIds": [ "aisb-mcq-fluency", "ot-capabilities-hallucination", "ot-capabilities-analysis", "ot-capabilities-fluency-not", "ot-capabilities-stats", "ot-cov-capabilities-1", "ot-cov-capabilities-2" ] }, { "id": "test-workflow-fit", "objectiveId": "aisb-obj-workflow-fit", "title": "Choosing a First AI Workflow", "mcqIds": [ "aisb-mcq-first-workflow", "ot-workflow-definition", "ot-workflow-worst", "ot-workflow-augment", "ot-workflow-best-task" ] } ] }, { "id": "aisb-l2", "title": "Prompting as Business Communication", "order": 1, "studyGuidePath": "/packs/ai-for-small-business/guides/l2-prompting.md", "sources": [ { "label": "OpenAI Academy — Small Business Resource Hub", "url": "https://academy.openai.com/en/public/resources/openai-academy-small-business-resource-hub-2026-06-03" }, { "label": "Ethan Mollick — Using AI Right Now: A Quick Guide", "url": "https://www.oneusefulthing.org/p/using-ai-right-now-a-quick-guide" } ], "parts": [ { "id": "aisb-l2p1-context", "title": "Give AI Context and a Clear Target", "order": 0, "blurb": "The practical anatomy of a useful request.", "studyGuideAnchor": "context", "itemIds": [ "aisb-fact-prompt-brief", "aisb-fact-examples", "aisb-fact-acceptance-criteria", "aisb-pair-task", "aisb-pair-context", "aisb-def-acceptance-criteria", "aisb-mcq-better-prompt", "aisb-proc-five-part-brief", "czr-ai-for-small-business-aisb-def-acceptance-criteria", "tf-d-ai-for-small-business-aisb-def-acceptance-criteria", "ot-brief-ingredients", "ot-brief-acceptance-def", "ot-brief-examples", "ot-brief-criteria-not" ] }, { "id": "aisb-l2p2-iteration", "title": "Iterate, Critique, and Reuse", "order": 1, "blurb": "Turn one-shot prompting into a dependable working conversation.", "studyGuideAnchor": "iteration", "itemIds": [ "aisb-fact-conversation", "aisb-fact-ask-questions", "aisb-fact-save-pattern", "aisb-pair-draft", "aisb-pair-critique", "aisb-def-prompt-library", "aisb-mcq-weak-output", "aisb-proc-iteration-loop", "czr-ai-for-small-business-aisb-def-prompt-library", "tf-df-ai-for-small-business-aisb-def-prompt-library", "ot-iter-library-def", "ot-iter-critique-pass", "ot-iter-save-pattern", "ot-iter-diagnose-promise", "ot-cov-iteration-1", "ot-cov-iteration-2" ] } ], "objectives": [ { "id": "aisb-obj-prompt-brief", "statement": "Write an AI brief with a clear task, relevant context, constraints, output format, and acceptance criteria.", "demonstrationIds": [ "aisb-demo-prompt-brief" ] }, { "id": "aisb-obj-iteration", "statement": "Improve weak output through critique, added examples, and reusable prompt-library assets rather than repeated one-shot prompting.", "demonstrationIds": [ "aisb-demo-iteration" ] } ], "demonstrations": [ { "id": "aisb-demo-prompt-brief", "label": "Building a useful prompt brief", "itemIds": [ "aisb-pair-task", "aisb-pair-context", "aisb-mcq-better-prompt", "czr-ai-for-small-business-aisb-def-acceptance-criteria" ], "requiredCorrect": 4 }, { "id": "aisb-demo-iteration", "label": "Critiquing and reusing prompts", "itemIds": [ "aisb-pair-draft", "aisb-pair-critique", "aisb-mcq-weak-output", "czr-ai-for-small-business-aisb-def-prompt-library", "ot-cov-iteration-1", "ot-cov-iteration-2" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-prompt-brief", "objectiveId": "aisb-obj-prompt-brief", "title": "Writing an AI Brief", "mcqIds": [ "aisb-mcq-better-prompt", "ot-brief-ingredients", "ot-brief-acceptance-def", "ot-brief-examples", "ot-brief-criteria-not" ] }, { "id": "test-iteration", "objectiveId": "aisb-obj-iteration", "title": "Iterate, Critique, Reuse", "mcqIds": [ "aisb-mcq-weak-output", "ot-iter-library-def", "ot-iter-critique-pass", "ot-iter-save-pattern", "ot-iter-diagnose-promise", "ot-cov-iteration-1", "ot-cov-iteration-2" ] } ] }, { "id": "aisb-l3", "title": "Research, Data, and Decisions", "order": 2, "studyGuidePath": "/packs/ai-for-small-business/guides/l3-research-decisions.md", "sources": [ { "label": "U.S. Census Bureau — AI use by U.S. businesses (May 2026)", "url": "https://www.census.gov/library/stories/2026/05/ai-use-businesses.html" }, { "label": "U.S. Small Business Administration — AI for small business", "url": "https://www.sba.gov/business-guide/manage-your-business/ai-small-business" } ], "parts": [ { "id": "aisb-l3p1-research", "title": "Research Current Markets and Competitors", "order": 0, "blurb": "Use AI to accelerate research without outsourcing judgment.", "studyGuideAnchor": "research", "itemIds": [ "aisb-fact-search-vs-memory", "aisb-fact-primary-sources", "aisb-fact-survey-definitions", "aisb-pair-primary-source", "aisb-pair-secondary-source", "aisb-def-triangulation", "aisb-mcq-market-claim", "aisb-proc-research-brief", "czr-ai-for-small-business-aisb-def-triangulation", "tf-d-ai-for-small-business-aisb-def-triangulation", "ot-research-primary", "ot-research-triangulate", "ot-research-when-search", "ot-research-definitions", "ot-cov-research-1", "ot-cov-research-2" ] }, { "id": "aisb-l3p2-data-decisions", "title": "Analyze Data and Make Better Decisions", "order": 1, "blurb": "Move from a spreadsheet to a decision without confusing observations with assumptions.", "studyGuideAnchor": "data-decisions", "itemIds": [ "aisb-fact-clean-data", "aisb-fact-observation-inference", "aisb-fact-scenarios", "aisb-pair-observation", "aisb-pair-inference", "aisb-def-sensitivity-analysis", "aisb-mcq-data-first", "aisb-proc-decision-memo", "czr-ai-for-small-business-aisb-def-sensitivity-analysis", "tf-df-ai-for-small-business-aisb-def-sensitivity-analysis", "ot-data-inference", "ot-data-sensitivity", "ot-data-scenarios", "ot-data-inspect" ] } ], "objectives": [ { "id": "aisb-obj-research", "statement": "Verify current market claims with primary sources and triangulation before using them in a business decision.", "demonstrationIds": [ "aisb-demo-research" ] }, { "id": "aisb-obj-data-decisions", "statement": "Separate observations from inferences and check data quality before recommending an action.", "demonstrationIds": [ "aisb-demo-data-decisions" ] } ], "demonstrations": [ { "id": "aisb-demo-research", "label": "Source quality and triangulation", "itemIds": [ "aisb-pair-primary-source", "aisb-pair-secondary-source", "aisb-mcq-market-claim", "czr-ai-for-small-business-aisb-def-triangulation", "ot-cov-research-1", "ot-cov-research-2" ], "requiredCorrect": 3 }, { "id": "aisb-demo-data-decisions", "label": "From data to a defensible decision", "itemIds": [ "aisb-pair-observation", "aisb-pair-inference", "aisb-mcq-data-first" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-research", "objectiveId": "aisb-obj-research", "title": "Verifying Market Claims", "mcqIds": [ "aisb-mcq-market-claim", "ot-research-primary", "ot-research-triangulate", "ot-research-when-search", "ot-research-definitions", "ot-cov-research-1", "ot-cov-research-2" ] }, { "id": "test-data-decisions", "objectiveId": "aisb-obj-data-decisions", "title": "Observations vs Inferences", "mcqIds": [ "aisb-mcq-data-first", "ot-data-inference", "ot-data-sensitivity", "ot-data-scenarios", "ot-data-inspect" ] } ] }, { "id": "aisb-l4", "title": "Marketing, Sales, and Customer Service", "order": 3, "studyGuidePath": "/packs/ai-for-small-business/guides/l4-customer-growth.md", "sources": [ { "label": "U.S. Small Business Administration — AI for small business", "url": "https://www.sba.gov/business-guide/manage-your-business/ai-small-business" }, { "label": "U.S. Copyright Office — Copyright and AI, Part 2", "url": "https://www.copyright.gov/ai/Copyright-and-Artificial-Intelligence-Part-2-Copyrightability-Report.pdf" } ], "parts": [ { "id": "aisb-l4p1-marketing", "title": "Create Marketing Without Losing Your Voice", "order": 0, "blurb": "Scale drafts and variations while keeping claims and brand judgment human-owned.", "studyGuideAnchor": "marketing", "itemIds": [ "aisb-fact-brand-context", "aisb-fact-content-atom", "aisb-fact-copyright", "aisb-pair-brand-voice", "aisb-pair-proof-point", "aisb-def-content-atom", "aisb-mcq-claim-review", "aisb-proc-content-workflow", "czr-ai-for-small-business-aisb-def-content-atom", "cz-ai-for-small-business-aisb-fact-copyright", "tf-d-ai-for-small-business-aisb-def-content-atom", "ot-marketing-brand-brief", "ot-marketing-content-atom", "ot-marketing-copyright", "ot-marketing-proof-point" ] }, { "id": "aisb-l4p2-sales-service", "title": "Support Sales and Customer Service", "order": 1, "blurb": "Respond faster while keeping promises, exceptions, and empathy under control.", "studyGuideAnchor": "sales-service", "itemIds": [ "aisb-fact-approved-knowledge", "aisb-fact-escalate", "aisb-fact-personalization", "aisb-pair-qualification", "aisb-pair-escalation", "aisb-def-escalation-rule", "aisb-mcq-chatbot-unknown", "aisb-proc-faq-assistant", "czr-ai-for-small-business-aisb-def-escalation-rule", "tf-df-ai-for-small-business-aisb-def-escalation-rule", "ot-cs-escalation-rule", "ot-cs-escalation-trigger", "ot-cs-grounding", "ot-cs-personalization", "ot-cov-customer-service-1", "ot-cov-customer-service-2" ] } ], "objectives": [ { "id": "aisb-obj-marketing", "statement": "Use AI to adapt marketing content while preserving brand voice and substantiating claims.", "demonstrationIds": [ "aisb-demo-marketing" ] }, { "id": "aisb-obj-customer-service", "statement": "Design customer-facing assistance that uses approved knowledge and escalates unknown, sensitive, or exceptional cases.", "demonstrationIds": [ "aisb-demo-customer-service" ] } ], "demonstrations": [ { "id": "aisb-demo-marketing", "label": "Brand-safe marketing and claims", "itemIds": [ "aisb-pair-brand-voice", "aisb-pair-proof-point", "aisb-mcq-claim-review" ], "requiredCorrect": 3 }, { "id": "aisb-demo-customer-service", "label": "Approved answers and escalation", "itemIds": [ "aisb-pair-qualification", "aisb-pair-escalation", "aisb-mcq-chatbot-unknown", "czr-ai-for-small-business-aisb-def-escalation-rule", "ot-cov-customer-service-1", "ot-cov-customer-service-2" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-marketing", "objectiveId": "aisb-obj-marketing", "title": "AI Marketing Content", "mcqIds": [ "aisb-mcq-claim-review", "ot-marketing-brand-brief", "ot-marketing-content-atom", "ot-marketing-copyright", "ot-marketing-proof-point" ] }, { "id": "test-customer-service", "objectiveId": "aisb-obj-customer-service", "title": "Customer Service & Escalation", "mcqIds": [ "aisb-mcq-chatbot-unknown", "ot-cs-escalation-rule", "ot-cs-escalation-trigger", "ot-cs-grounding", "ot-cs-personalization", "ot-cov-customer-service-1", "ot-cov-customer-service-2" ] } ] }, { "id": "aisb-l5", "title": "Operations, Automation, and Agents", "order": 4, "studyGuidePath": "/packs/ai-for-small-business/guides/l5-operations-automation.md", "sources": [ { "label": "U.S. Small Business Administration — AI for small business", "url": "https://www.sba.gov/business-guide/manage-your-business/ai-small-business" }, { "label": "NIST — Generative AI Profile (NIST AI 600-1)", "url": "https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf" } ], "parts": [ { "id": "aisb-l5p1-operations", "title": "Document Operations and Repetitive Work", "order": 0, "blurb": "Turn owner knowledge into usable procedures and team support.", "studyGuideAnchor": "operations", "itemIds": [ "aisb-fact-capture-process", "aisb-fact-exceptions", "aisb-fact-meeting-action", "aisb-pair-happy-path", "aisb-pair-exception-path", "aisb-def-sop", "aisb-mcq-sop-source", "aisb-proc-build-sop", "czr-ai-for-small-business-aisb-def-sop", "tf-df-ai-for-small-business-aisb-def-sop", "ot-sop-exception-path", "ot-sop-not-complete", "ot-sop-hidden-decisions", "ot-sop-test-person" ] }, { "id": "aisb-l5p2-automation", "title": "Automate Carefully and Introduce Agents", "order": 1, "blurb": "Connect tools only after the underlying workflow is understood.", "studyGuideAnchor": "automation", "itemIds": [ "aisb-fact-trigger-action", "aisb-fact-agent-permission", "aisb-fact-logs", "aisb-pair-automation", "aisb-pair-agent", "aisb-def-approval-gate", "aisb-mcq-first-agent", "aisb-proc-automate-safely", "czr-ai-for-small-business-aisb-def-approval-gate", "tf-d-ai-for-small-business-aisb-def-approval-gate", "ot-auto-agent-vs-rule", "ot-auto-approval-gate", "ot-auto-least-privilege", "ot-auto-rule-enough" ] } ], "objectives": [ { "id": "aisb-obj-sop", "statement": "Capture an experienced worker's real process, including happy paths and exceptions, as a testable standard operating procedure.", "demonstrationIds": [ "aisb-demo-sop" ] }, { "id": "aisb-obj-automation", "statement": "Distinguish rule-based automation from an AI agent and place human approval before consequential actions.", "demonstrationIds": [ "aisb-demo-automation" ] } ], "demonstrations": [ { "id": "aisb-demo-sop", "label": "Documenting a real operating process", "itemIds": [ "aisb-pair-happy-path", "aisb-pair-exception-path", "aisb-mcq-sop-source" ], "requiredCorrect": 3 }, { "id": "aisb-demo-automation", "label": "Automation, agents, and approval gates", "itemIds": [ "aisb-pair-automation", "aisb-pair-agent", "aisb-mcq-first-agent", "czr-ai-for-small-business-aisb-def-approval-gate" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-sop", "objectiveId": "aisb-obj-sop", "title": "Standard Operating Procedures", "mcqIds": [ "aisb-mcq-sop-source", "ot-sop-exception-path", "ot-sop-not-complete", "ot-sop-hidden-decisions", "ot-sop-test-person" ] }, { "id": "test-automation", "objectiveId": "aisb-obj-automation", "title": "Automation and Agents", "mcqIds": [ "aisb-mcq-first-agent", "ot-auto-agent-vs-rule", "ot-auto-approval-gate", "ot-auto-least-privilege", "ot-auto-rule-enough" ] } ] }, { "id": "aisb-l6", "title": "High-Stakes Work: Money, Contracts, and People", "order": 5, "studyGuidePath": "/packs/ai-for-small-business/guides/l6-high-stakes.md", "sources": [ { "label": "EEOC and DOJ — AI and disability discrimination in employment", "url": "https://www.eeoc.gov/newsroom/us-eeoc-and-us-department-justice-warn-against-disability-discrimination" }, { "label": "NIST — AI Risk Management Framework", "url": "https://airc.nist.gov/airmf-resources/airmf/" } ], "parts": [ { "id": "aisb-l6p1-finance-legal", "title": "Use AI Around Money, Contracts, and Taxes", "order": 0, "blurb": "Accelerate preparation and review without delegating professional judgment.", "studyGuideAnchor": "finance-legal", "itemIds": [ "aisb-fact-first-pass", "aisb-fact-reconcile", "aisb-fact-professional", "aisb-pair-first-pass", "aisb-pair-final-decision", "aisb-def-material-term", "aisb-mcq-lease", "aisb-proc-high-stakes-review", "czr-ai-for-small-business-aisb-def-material-term", "tf-df-ai-for-small-business-aisb-def-material-term", "ot-high-stakes-material-term", "ot-high-stakes-reconcile", "ot-high-stakes-tax", "ot-high-stakes-responsibility" ] }, { "id": "aisb-l6p2-people", "title": "Support Hiring and People Work Fairly", "order": 1, "blurb": "Use AI for administration while keeping employment decisions accountable and lawful.", "studyGuideAnchor": "people", "itemIds": [ "aisb-fact-employment-law", "aisb-fact-accommodation", "aisb-fact-human-decision", "aisb-pair-objective-criterion", "aisb-pair-subjective-judgment", "aisb-def-adverse-impact", "aisb-mcq-resume-screen", "aisb-proc-hiring-aid", "czr-ai-for-small-business-aisb-def-adverse-impact", "tf-d-ai-for-small-business-aisb-def-adverse-impact", "ot-people-adverse-impact", "ot-people-eeoc", "ot-people-human-review", "ot-people-protected", "ot-cov-people-1", "ot-cov-people-2" ] } ], "objectives": [ { "id": "aisb-obj-high-stakes", "statement": "Use AI for preparation around contracts while reserving material decisions for a qualified human.", "demonstrationIds": [ "aisb-demo-high-stakes" ] }, { "id": "aisb-obj-people", "statement": "Use AI for hiring administration without delegating employment judgment or ignoring adverse-impact risk.", "demonstrationIds": [ "aisb-demo-people" ] } ], "demonstrations": [ { "id": "aisb-demo-high-stakes", "label": "AI preparation versus professional judgment", "itemIds": [ "aisb-pair-first-pass", "aisb-pair-final-decision", "aisb-mcq-lease", "czr-ai-for-small-business-aisb-def-material-term" ], "requiredCorrect": 4 }, { "id": "aisb-demo-people", "label": "Fair and accountable people decisions", "itemIds": [ "aisb-pair-objective-criterion", "aisb-pair-subjective-judgment", "aisb-mcq-resume-screen", "czr-ai-for-small-business-aisb-def-adverse-impact", "ot-cov-people-1", "ot-cov-people-2" ], "requiredCorrect": 3 } ], "objectiveTests": [ { "id": "test-high-stakes", "objectiveId": "aisb-obj-high-stakes", "title": "Contracts, Finance & Taxes", "mcqIds": [ "aisb-mcq-lease", "ot-high-stakes-material-term", "ot-high-stakes-reconcile", "ot-high-stakes-tax", "ot-high-stakes-responsibility" ] }, { "id": "test-people", "objectiveId": "aisb-obj-people", "title": "Fair Hiring with AI", "mcqIds": [ "aisb-mcq-resume-screen", "ot-people-adverse-impact", "ot-people-eeoc", "ot-people-human-review", "ot-people-protected", "ot-cov-people-1", "ot-cov-people-2" ] } ] }, { "id": "aisb-l7", "title": "Privacy, Security, and Responsible Use", "order": 6, "studyGuidePath": "/packs/ai-for-small-business/guides/l7-governance.md", "sources": [ { "label": "Federal Trade Commission — Privacy and security guidance", "url": "https://www.ftc.gov/business-guidance/privacy-security" }, { "label": "NIST — AI Risk Management Framework", "url": "https://airc.nist.gov/airmf-resources/airmf/" }, { "label": "OpenAI — Business data privacy, security, and compliance", "url": "https://openai.com/business-data/" } ], "parts": [ { "id": "aisb-l7p1-data", "title": "Protect Customer and Business Data", "order": 0, "blurb": "Match the tool and workflow to the sensitivity of the information.", "studyGuideAnchor": "data", "itemIds": [ "aisb-fact-data-minimization", "aisb-fact-plan-controls", "aisb-fact-business-data", "aisb-pair-public-data", "aisb-pair-confidential-data", "aisb-def-data-retention", "aisb-mcq-customer-list", "aisb-proc-data-check", "czr-ai-for-small-business-aisb-def-data-retention", "cz-ai-for-small-business-aisb-fact-business-data", "tf-d-ai-for-small-business-aisb-def-data-retention", "ot-data-protection-minimization", "ot-data-protection-plan-controls", "ot-data-protection-public", "ot-data-protection-retention", "ot-cov-data-protection-1", "ot-cov-data-protection-2", "ot-cov-data-protection-3", "ot-cov-data-protection-4" ] }, { "id": "aisb-l7p2-governance", "title": "Build Human Oversight and Quality Controls", "order": 1, "blurb": "Scale AI use with risk tiers, testing, logs, and incident response.", "studyGuideAnchor": "governance", "itemIds": [ "aisb-fact-risk-tier", "aisb-fact-nist-functions", "aisb-fact-incidents", "aisb-pair-reversible", "aisb-pair-irreversible", "aisb-def-human-in-loop", "aisb-mcq-review-level", "aisb-proc-preflight", "czr-ai-for-small-business-aisb-def-human-in-loop", "tf-df-ai-for-small-business-aisb-def-human-in-loop", "ot-governance-consequential", "ot-governance-human-loop", "ot-governance-incident", "ot-governance-risk-tier" ] } ], "objectives": [ { "id": "aisb-obj-data-protection", "statement": "Classify business information, minimize shared data, and check vendor retention and training controls before use.", "demonstrationIds": [ "aisb-demo-data-protection" ] }, { "id": "aisb-obj-governance", "statement": "Match oversight to consequence using risk tiers and human review.", "demonstrationIds": [ "aisb-demo-governance" ] } ], "demonstrations": [ { "id": "aisb-demo-data-protection", "label": "Data classification and minimization", "itemIds": [ "aisb-pair-public-data", "aisb-pair-confidential-data", "aisb-mcq-customer-list", "aisb-def-data-retention", "ot-cov-data-protection-1", "ot-cov-data-protection-2", "ot-cov-data-protection-3", "ot-cov-data-protection-4" ], "requiredCorrect": 3 }, { "id": "aisb-demo-governance", "label": "Risk-based oversight and controls", "itemIds": [ "aisb-pair-reversible", "aisb-pair-irreversible", "aisb-mcq-review-level", "aisb-def-human-in-loop" ], "requiredCorrect": 4 } ], "objectiveTests": [ { "id": "test-data-protection", "objectiveId": "aisb-obj-data-protection", "title": "Data Protection & Classification", "mcqIds": [ "aisb-mcq-customer-list", "ot-data-protection-minimization", "ot-data-protection-plan-controls", "ot-data-protection-public", "ot-data-protection-retention", "ot-cov-data-protection-1", "ot-cov-data-protection-2", "ot-cov-data-protection-3", "ot-cov-data-protection-4" ] }, { "id": "test-governance", "objectiveId": "aisb-obj-governance", "title": "Governance & Human Oversight", "mcqIds": [ "aisb-mcq-review-level", "ot-governance-consequential", "ot-governance-human-loop", "ot-governance-incident", "ot-governance-risk-tier" ] } ] }, { "id": "aisb-l8", "title": "A 30-Day Implementation Playbook", "order": 7, "studyGuidePath": "/packs/ai-for-small-business/guides/l8-implementation.md", "sources": [ { "label": "OpenAI Academy — Small Business Resource Hub", "url": "https://academy.openai.com/en/public/resources/openai-academy-small-business-resource-hub-2026-06-03" }, { "label": "NIST — AI Risk Management Framework", "url": "https://airc.nist.gov/airmf-resources/airmf/" }, { "label": "U.S. Small Business Administration — AI for small business", "url": "https://www.sba.gov/business-guide/manage-your-business/ai-small-business" } ], "parts": [ { "id": "aisb-l8p1-pilot", "title": "The 30-Day Pilot", "order": 0, "blurb": "Recognize the building blocks of a measured first implementation.", "studyGuideAnchor": "pilot", "itemIds": [ "aisb-fact-one-pilot", "aisb-fact-baseline", "aisb-fact-acceptance-rate", "aisb-concept-task-inventory", "aisb-concept-baseline", "aisb-concept-pilot", "aisb-concept-parallel-run", "aisb-concept-acceptance-criteria", "aisb-concept-roi", "aisb-mcq-pilot-success", "cz-ai-for-small-business-aisb-fact-one-pilot", "ot-pilot-define", "ot-pilot-baseline", "ot-pilot-parallel", "ot-pilot-inventory", "ot-cov-pilot-1", "ot-cov-pilot-2" ] }, { "id": "aisb-l8p2-operating-system", "title": "Your Small-Business AI Operating System", "order": 1, "blurb": "Recognize the controls that keep useful AI use repeatable and accountable.", "studyGuideAnchor": "operating-system", "itemIds": [ "aisb-fact-policy-light", "aisb-fact-owner", "aisb-fact-quarterly-review", "aisb-concept-ai-policy", "aisb-concept-prompt-library", "aisb-concept-approval-gate", "aisb-concept-audit-log", "aisb-concept-escalation-rule", "aisb-concept-kill-switch", "aisb-mcq-operating-system", "ot-os-policy", "ot-os-killswitch", "ot-os-escalation", "ot-os-auditlog" ] } ], "objectives": [ { "id": "aisb-obj-pilot", "statement": "Plan a measured 30-day pilot with a baseline, acceptance criteria, parallel run, and evidence-based expansion decision.", "demonstrationIds": [ "aisb-demo-pilot" ] }, { "id": "aisb-obj-operating-system", "statement": "Recognize the policies, approval gates, and audit logs that make AI use dependable.", "demonstrationIds": [ "aisb-demo-operating-system" ] } ], "demonstrations": [ { "id": "aisb-demo-pilot", "label": "Designing and evaluating a 30-day pilot", "itemIds": [ "aisb-concept-task-inventory", "aisb-concept-baseline", "aisb-concept-parallel-run", "aisb-mcq-pilot-success", "ot-cov-pilot-1", "ot-cov-pilot-2" ], "requiredCorrect": 3 }, { "id": "aisb-demo-operating-system", "label": "The small-business AI operating system", "itemIds": [ "aisb-concept-ai-policy", "aisb-concept-approval-gate", "aisb-concept-audit-log", "aisb-mcq-operating-system" ], "requiredCorrect": 4 } ], "objectiveTests": [ { "id": "test-pilot", "objectiveId": "aisb-obj-pilot", "title": "The 30-Day Pilot", "mcqIds": [ "aisb-mcq-pilot-success", "ot-pilot-define", "ot-pilot-baseline", "ot-pilot-parallel", "ot-pilot-inventory", "ot-cov-pilot-1", "ot-cov-pilot-2" ] }, { "id": "test-operating-system", "objectiveId": "aisb-obj-operating-system", "title": "The AI Operating System", "mcqIds": [ "aisb-mcq-operating-system", "ot-os-policy", "ot-os-killswitch", "ot-os-escalation", "ot-os-auditlog" ] } ] } ] }