Verifying AI output before you use it
For anyone who uses an approved AI assistant to draft documents, emails, procedures or scripts that other people will rely on. Builds on the course 'Getting reliable results from AI', which you should complete beforehand; this one is the hands-on part. You'll run a verification pass on a draft, trace citations to real sources, check figures against the original data, check product steps against current vendor documentation for your version, treat AI-suggested commands and code as untested, watch for instructions hidden in content you ask AI to summarize, and record what you checked. Your organization's AI policy governs which tools you may use.
- Level
- Intermediate
- Length
- About 70 minutes
- Contents
- 5 lessons · final exam
- Status
- Published · updated 10 Oct 2026
Skills you'll practise
- Identify every checkable claim in an AI draft: citations, quotes, figures, names, dates, product steps and commands
- Trace a citation to its source and decide whether it exists and supports the claim as worded
- Calculate or look up figures from the original data and correct the draft
- Choose the right source to check a product procedure: current vendor documentation for the version you run
- Write a verification log that records what was checked, against what, and the outcome, and report hidden instructions found in AI summaries
Course outline
- 1.The verification pass: list what can be checkedLesson · 12 min
- 2.Citations and quotes: does it exist, and does it say that?Lesson · 14 min
- 3.Figures: look up and calculate from the original data, then correct the draftLesson · 7 min
- 4.Product procedures and commands: check current vendor documentation for your versionLesson · 9 min
- 5.Writing a verification log, and hidden instructions in summariesLesson · 12 min
- 6.Verifying AI output before you use it: knowledge checkKnowledge check · 14 questions
- 7.Verifying AI output before you use it: practical exerciseKnowledge check · 1 question
- 8.Final exam10 questions · passing it completes the course, so people who already know the material can test out
Sources it draws on
The lessons and questions are written from these references, so learners can go back to the original.
- NIST AI 600-1: Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (July 2024)
- NIST publication record for AI 600-1 (title, date, authors, abstract)
- UK NCSC: AI and cyber security: what you need to know (February 2024)
- OWASP Top 10 for LLM Applications 2025: LLM09 Misinformation
- OWASP Top 10 for LLM Applications 2025: LLM05 Improper Output Handling
See it with your own jobs and topics
Tell us about your team and we'll walk you through setup, from choosing jobs to your first skills check.