AI glossary / Safety and trust
Guardrails
Rules and filters built around an AI to stop it producing harmful or inappropriate responses.
More words in Safety and trust
- Human in the loopKeeping a person involved to check or approve what the AI does, especially for decisions that matter.
- JailbreakA cleverly worded request designed to trick an AI into ignoring its safety rules.
- Model cardA short fact sheet published with a model that explains what it's for, how it was tested and where it falls short.
- Model specA public document in which an AI company sets out how its assistant is supposed to behave.
- OverrelianceTrusting AI output without checking it, or letting your own skills fade because the AI always does it for you.