AI safety guides
Use AI with sensible checks and clear boundaries
AI can be useful without being automatically right, private or suitable for every decision. These guides show what to check, what to keep out and where human judgement still matters.
Choose a guide
Practical help for this part of AI
Each guide focuses on one question or decision, with a clear next step you can use.
A useful place to begin: Start with the information you should never share, then learn how to check a confident answer.
What Should You Never Share With an AI Chatbot?
Know which personal, account and workplace information should stay out of a general AI chatbot.
Open guide → Check the answerAI Sounds Confident. That Does Not Mean It Is Right
AI can sound certain even when it is wrong. Learn what needs checking and when to find better evidence.
Open guide → Understand memoryAI Memory Explained: What It Remembers and What It Does Not
Understand chat context, saved memory, past conversations and projects without assuming AI remembers everything.
Open guide → Keep human judgementShould AI Make the Decision or Help You Make It?
Use AI to organise evidence and compare options without handing over responsibility for the final choice.
Open guide →Your next step
Want a practical checking process?
The Quick Guides include short methods for checking AI output, setting task boundaries and keeping important work under review.
Browse Quick GuidesMore ways to learn
