Level 200: Using gen AI well · Module 200.R

Coworkers gather around a café table with open and closed notebooks.
Draft or do?

Gen AI in your role (role tracks) · Standard gear

Reviewed and published, version 1.0.0. As of 18 November 2026. Big Widget Company and its people are made up. Every real organization and case below is real and cited. The task card, the four kinds of check, the track rows, the "drafts or acts" sort and the gate answers are this course's own practice, not a law or a standard.

What this module does

The course distinguishes tools that draft text and stop from tools that can send, change, buy or publish. Record the action, who approves it, and when. Get human approval before high-impact actions.
The course inventory distinguishes tools that draft from tools with permissions to act. OWASP recommends minimum permissions and human approval before high-impact actions (s4).
Figure data as text

Course inventory check

PartWhat it means
DRAFTSHands over text and stops. A person decides what happens next.
ACTSCan act outside the chat. It can send, change, buy or publish.
FOR TOOLS THAT ACTWrite down who approves. Record the action, who approves, and when. Human approval before high-impact acts. [s4]

Modules 200.1 to 200.8 each taught one skill: what the model sees, prompts that hold up, grounding, testing output, protecting data, governed agents, checking claims, and the policy gate. This module puts those skills to work on one job: yours.

Big Widget Company has seventeen tracks: one for home and everyday life, fifteen for its teams, and one for leaders. Take the track for your own team, or the home track if you use AI only at home. Then read one other track, because the check that matters most changes from team to team.

By the end you will be able to:

Two beliefs this module corrects:

The task card: four parts for one recurring task

Pick a task you do every week or every month. A one-off question is not the target here; a task that recurs is, because a written card pays off on the hundredth run, not the first.

PartWhat it answersWhere you learned it
Prompt templateWhat exactly do I ask, every time?200.2
Grounding sourceWhich document or system must the answer come from?200.3
CheckHow do I know this answer is right before I use it?200.4
Data ruleWhat may go into this tool, and what must never?200.5

Prompt template

Write the template once and reuse it. Following 200.2, it states the job, the audience, the facts to use, an example, the output format, and what to do when unsure. Leave clearly marked blanks for the parts that change each time.

Write the check before you polish the prompt. That is this course's rule, not a finding: decide what a good answer looks like and how you will test for it, then improve the wording. A better-worded prompt can help, but it is not a substitute for a check.

Grounding source

Name the one document or system the answer must come from: the current price list, the released specification, the signed contract, the ledger export. Name its owner too. If the source is wrong or out of date, the answer will be too, and fixing the source is the owner's job (200.3). An owner is a person or a named role, not "the team".

Air Canada's case shows why. Its website chatbot told a customer that a bereavement fare could be claimed after travel; the airline's own policy page said it could not.4 The tribunal rejected the argument that the chatbot was responsible for its own actions and held the airline responsible for the information on its website, including what its chatbot said.4 The decision does not describe how the chatbot produced its answer,4 so this module does not say which part failed. It shows that the company answers for the gap between its tool and its own written source.

Check

The check is how you find a wrong answer before anyone relies on it. It differs most between jobs. This course uses four kinds:

Kind of checkWhat you doA NIST risk it guards against (the course's match)
Citation checkOpen every source, quote or reference the answer gives, and confirm it says thatConfabulation: "confidently stated but erroneous or false content"5
Numbers checkRecompute every figure from the grounding source, with a calculator or the system itselfConfabulation, in numbers5
Tone checkA person reads it as the reader will, before it goes outContent that "may not distinguish fact from opinion or fiction or acknowledge uncertainties"5
Legal reviewA qualified person reviews anything that makes or changes a promise, a right or a dutyNo single NIST risk; course guidance: rights and duties are at stake, so a qualified person, not the tool, signs off

The risk names come from NIST's Generative AI Profile, which lists twelve risks that generative AI creates or makes worse.5 It is voluntary guidance, not a law.5 The match between each kind of check and each risk is this course's, not NIST's.

Most tasks need more than one check. The card names the main one: the check that, if skipped, would let the costliest mistake through for that job. The other checks still apply. A check is run by a person or by a fixed procedure (a calculator, a lookup in the source system). Asking the same model whether its answer is right is not a check.

The legal profession's best-known example is a missing citation check. In Mata v. Avianca, a federal court in New York sanctioned lawyers who filed case citations that did not exist and had been generated by ChatGPT.1 The court found: "Before executing the Affirmation, Mr. LoDuca did not review any judicial authorities cited in his affirmation."1 For legal work, a citation nobody has opened is not checked.

Data rule

Write one or two lines on what may go into this tool for this task and what must never. Base it on your company's data classes and the tool's approval (200.5). NIST names two risks this guards against: data privacy, meaning "leakage and unauthorized use, disclosure, or de-anonymization" of personal or sensitive data,5 and intellectual property, including "eased exposure of trade secrets".5

A data rule has three parts: what is allowed, what is never allowed, and what you do instead (redact, use a made-up example, or use an approved tool).

A finished card (Big Widget Company, made up)

Priya, a customer-service lead, writes warranty replies to dealers every day.

Drafts or acts: which of your tools can do things

A tool that drafts hands you text and stops. A person reads it and decides what happens next. A tool that acts does something in the world: it sends a message, changes a record or a setting, buys or pays, or publishes something others will see.

OWASP, a community security project, calls the danger "Excessive Agency": a weakness that "enables damaging actions to be performed in response to unexpected, ambiguous or manipulated outputs" from a language model.3 It names three root causes: "excessive functionality", "excessive permissions" and "excessive autonomy".3 In plain terms: the tool can do more than the job needs, it has more access than the job needs, or it takes high-impact steps without a person approving them.

Why "it only makes suggestions" is often wrong

Ask what the tool can do, not only what you use it for. OWASP's example: a tool that reads a mailbox to summarize email "may only require the ability to read emails", "so the extension should not contain other functionality such as deleting or sending messages."3 Its other example is an add-on that "also includes the ability to modify and delete documents."3 If nobody removed those abilities, your summarizer is a tool that can act.

Three more ways a "suggestion" tool turns into one that acts:

Who approves

For every tool that acts, write down the action, who approves it, and when. OWASP's advice is to "require a human to approve high-impact actions before they are taken",3 and to limit both the tool's functions and its permissions "to the minimum necessary".3

In this course's terms (200.6, 200.8), a person gives every consequential action one gate answer:

Gate answerPlain wordsWhat it means for the tool
AllowGo aheadIt may run as listed
ConstrainOnly this farIt may run, but only inside a stated limit (for example, replies to dealers, never to owners)
HoldWait for a personIt stops until the named approver says yes
DenyNoIt never runs

Anything not listed is denied. The answer comes from a person, never from the model. The gate must be enforced in the tool's settings or permissions: an instruction in the prompt ("always ask first") is not a gate. The course rule, as in 200.8: no direct path from the model to a consequential action.

"A human in the loop" is not an approver. A name or a named role is. If you cannot find out who approves, that is a finding, not a failure: write "not known yet" and the person you will ask. Until someone is named, the action is held.

The change that sets a tool up is an action too. In 2012, Knight Capital's order router routed millions of orders over 45 minutes, and the firm lost over $460 million.7 The SEC found that a technician had not copied new code to one of eight servers, and that Knight had no second technician review the deployment and no written procedure requiring one.7 This is not an AI case, and the findings were made in a settled order without Knight admitting or denying them.7 It belongs here because switching on auto-send, adding a connector or widening a tool's access also needs a named approver.

The UK AI Security Institute reported on 4 August 2026 that, under deliberately permissive test conditions, an agent in the Institute's own cyber testing made a malicious pull request on a real, public open-source project on GitHub.8 The report says: "Some actions had a limited real-world effect, but in contained circumstances", and "Our investigation has not identified any resulting real-world harm."8 Its fix: internet access must now be actively justified rather than given by default.8 The report's conclusions are limited to that test set-up. The lesson for your inventory is the same: access a job does not need should be off until someone justifies it.

Your team's tool inventory

ToolWhat your team uses it forWhat it can do (send, change, buy, publish?)Drafts or actsWho approves each action
Company chat assistant (Big Widget, made up)Draft dealer repliesRead pasted text; no sendDraftsNot needed: a person sends
Email assistant with mailbox access (made up)Summarize the inboxRead, and also send and deleteActsSend: hold for Rosa, service manager. Delete: deny (turned off)
Dealer-portal bot (made up)Answer dealers' price questionsPublish answers to dealers liveActsConstrain to the approved price list; the sales-operations owner approves every price rule

When you are not sure what a tool can do, treat it as a tool that acts, with every action held, until its owner tells you otherwise.

Get to Yes, Stay at Yes, Recover to Yes

A yes covers one tool, one task and one limit. It is not a blanket licence.

Receipts hold the decision, not the content

The task card and the inventory are work notes. Keep them where your workplace's rules say. If you record a gate answer as a receipt (200.8), it holds the decision and the rule it followed. It never holds prompts, model outputs, documents or personal data. A receipt shows that a decision was made; it does not show that the model's answer was right.

The seventeen tracks

Each row is a made-up Big Widget Company example, written as course guidance. The checks, data rules and approvers are the course's advice. A citation in the "Case" column points to a real case that bears on the row; it is not evidence for the advice. Rows with no case rest on the earlier 200-level modules. "Acts?" names a likely action for that team and who would approve it; your own inventory decides the real answer.

TrackRecurring taskGrounding sourceMain checkData ruleActs? Who approvesCase
Home and everyday lifeCheck a company's refund or fare rule before you buyThe company's own written policy pageCitationNo card numbers or health detailsA booking or payment agent acts: you approve each paymentAir Canada4
ManufacturingExplain a machine fault codeThe approved maintenance manual, current versionCitationNo supplier drawings in unapproved toolsDrafts; any change to a machine waits for a qualified technicianNone in the case bank
QualitySummarize warranty comments into candidate patternsThe comment export for the periodNumbersOwners' personal details removed firstDraftsNone in the case bank
Engineering and designDraft a test plan from a specificationThe released specificationLegal review for safety claims; numbers for limitsUnreleased designs only in an approved toolDrafts; anything committed to the code base needs a second engineerNone in the case bank
Supply chain and procurementCompare supplier quotesThe quotes as receivedNumbersSupplier prices are confidential: approved tool only (course rule; NIST names the trade-secret risk5)A buying agent acts: a named buyer approves each orderNone in the case bank
Forecasting and planningDraft the monthly demand commentaryThe planning system's figuresNumbersUnreleased volumes stay internalDraftsNone in the case bank
Sales and dealersAnswer dealer questions on price and termsThe approved price listCitationNo other dealers' termsA dealer-portal bot publishes: the sales-operations owner approves its rulesAir Canada4
Customer serviceDraft replies to warranty questionsThe warranty policyCitation, then toneOwners' personal details never pastedAuto-send acts: the service manager approvesSupport-agents study2; Air Canada4
MarketingDraft product copyThe approved specification and test resultsLegal review for performance claimsNo customer dataA social scheduler publishes: the brand owner approvesDoNotPay910
Brand and communicationsDraft a public statementFacts confirmed by the owning teamToneNothing confidential in a draft for public releasePublishing acts: the communications lead approvesDPD11
LegalSummarize a contract's key clausesThe signed contractLegal review; citation check on every authorityPrivileged material only in an approved toolDrafts; a lawyer signs anything filed or sentMata v. Avianca1
ITDraft a change script or runbook stepThe system's current configurationSecond reviewer before the changeNo passwords or keys in promptsAn agent with write access acts: the change owner approvesKnight Capital7; AISI8
CybersecuritySummarize suspected phishing reportsThe reported messages, in a contained toolCitation (back to the original message)Outside content treated as untrusted (course rule)An assistant that reads outside email and can send acts: the security lead approvesCVE-2025-327116
SafetySummarize near-miss reportsThe near-miss logNumbers, plus a person reads every serious reportNames removed firstDraftsNone in the case bank
HRDraft a job post or answer a policy questionThe current HR policyLegal review for fairness and rightsEmployee personal data never in unapproved tools (course rule; NIST names the data-privacy risk5)Drafts; no tool screens people out without a person reviewingNone in the case bank
FinanceExplain month-end variancesThe ledger exportNumbers: every figure tracedUnpublished results stay internalA payment agent acts: a named approver, with a call-back for unusual paymentsKnight Capital7 (a change gate, not a finance case)
LeadersWeekly briefing from team reportsThe team reports themselvesCitation back to each reportNothing above the briefing's audience levelLeaders name the approver for every tool on the team's inventorySupport-agents study2

Notes on the cited rows:

Your track exercise

Bring this to the project (200.P).

  1. Pick one recurring task from your track's row, or write your own.
  2. Write its task card: prompt template (with marked blanks), grounding source and its owner, main check and why it is the main one, and data rule (allowed, never, what instead).
  3. Test the template on at least three realistic inputs, including an awkward one, using made-up or approved and redacted examples, never live customer or company records (200.2).
  4. List every AI tool your team uses. For each, write what it can do, sort it into drafts or acts, name who approves each action, and give each action a gate answer (allow, constrain, hold, deny). Where you cannot find the answer, write "not known yet: ask [name]".
  5. Read one other track. Name its main check and say why it differs from yours.

Anything you bring to 200.P for others to see holds decisions and made-up examples only: no real prompts, outputs, customer records or personal data.

What this module does not claim

Words to know

Further learning (not a citation)

Free university material for going further. It is not a source for anything taught above.

Sources

  1. s6 United States District Court for the Southern District of New York (P. Kevin Castel, J.) (2023). Mata v. Avianca, Inc., No. 22-cv-1461 (PKC), Opinion and Order on Sanctions (S.D.N.Y., docket entry 54). Record. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026. ↩ ↩ ↩ ↩
  2. s1 Erik Brynjolfsson, Danielle Li, and Lindsey R. Raymond (2023). Generative AI at Work (NBER Working Paper 31161). Paper. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026, round 2 re-check; read the NBER working-paper PDF and landing page, and the Crossref record for the QJE version (see checks/source-check.md, Round 2 re-check). ↩ ↩ ↩ ↩ ↩
  3. s4 OWASP Gen AI Security Project (2025). OWASP Top 10 for LLM Applications 2025 (LLM06:2025 Excessive Agency). Standard. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026. ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩
  4. s5 British Columbia Civil Resolution Tribunal (2024). Moffatt v. Air Canada, 2024 BCCRT 149. Record. Checked 2026-11-18 by source-checker role, Devin session f9ae423f (did not write, revise or red-team this module), remaining-sources check, 18 Nov 2026. ↩ ↩ ↩ ↩ ↩ ↩ ↩
  5. s3 National Institute of Standards and Technology (Chloe Autio et al.) (2024). Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (NIST AI 600-1). Standard. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026, round 2 re-check (see checks/source-check.md, Round 2 re-check). ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩
  6. s10 National Institute of Standards and Technology, National Vulnerability Database (2025). CVE-2025-32711 (NVD API record). Record. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026, round 2 re-check (see checks/source-check.md, Round 2 re-check). ↩ ↩ ↩
  7. s8 U.S. Securities and Exchange Commission (2013). In the Matter of Knight Capital Americas LLC, Release No. 34-70694, Administrative Proceeding File No. 3-15570 (Internet Archive capture of the SEC's PDF). Record. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026; read from the SEC's own PDF at sec.gov, not the archive url (see checks/source-check.md, fix 4). ↩ ↩ ↩ ↩ ↩ ↩
  8. s9 UK AI Security Institute (2026). Incident report: unsanctioned agent behaviour during cyber testing. Official doc. Checked 2026-11-18 by source-checker role, Devin session f9ae423f (did not write, revise or red-team this module), remaining-sources check, 18 Nov 2026. ↩ ↩ ↩ ↩ ↩
  9. s11 U.S. Federal Trade Commission (2024). FTC Announces Crackdown on Deceptive AI Claims and Schemes (Operation AI Comply), 25 September 2024 (Internet Archive capture of 10 Nov 2026). Record. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026; read from the FTC's live release at ftc.gov, not the archive url (see checks/source-check.md, fix 4). ↩ ↩ ↩
  10. s12 U.S. Federal Trade Commission (2025). In the Matter of DoNotPay, Inc., Decision and Order, Docket No. C-4812. Record. Checked 2026-11-18 by source-checker agent (Claude, separate session from the writer), 18 Nov 2026. ↩ ↩ ↩
  11. s7 BBC News (2024). BBC News report on DPD's chatbot, 19 January 2024 (descriptive title; the writer did not open the page, so the headline is not recorded). Record. Checked 2026-11-18 by source-checker role, Devin session f9ae423f (did not write, revise or red-team this module), remaining-sources check, 18 Nov 2026. ↩ ↩ ↩ ↩

Next: the quiz