AI glossary

A

Safety AI regulation Laws governing how AI may be built and used, such as the EU AI Act, which sets obligations by risk level. Safety AI safety Work to prevent AI systems from causing harm, through misuse, accidents or behaviour their developers did not intend.

B

Safety Bias Systematic unfairness in a model's outputs, often inherited from patterns in its training data.

C

Safety Content moderation Detecting and filtering harmful content such as hate, violence or sexual material, in text or images.

D

Safety Deepfake Realistic synthetic media, usually video or audio, that shows someone saying or doing something they did not.

E

Safety Explainability How well people can understand why an AI system produced a given output.

G

Safety Guardrails Checks and rules around an AI system that block unsafe inputs or outputs and keep it within its intended scope.

J

Safety Jailbreak A prompt crafted to get a model to ignore its safety rules and produce content it would normally refuse.

P

Safety Prompt injection An attack where instructions hidden in content the model reads, such as a web page, email or document, try to hijack its behaviour.

R

Safety Red teaming Deliberately attacking an AI system to find its failures and harmful behaviours before release. Safety Responsible AI Developing and using AI in ways that are fair, transparent, accountable, private and safe.

S

Safety System card (model card) A document a provider publishes with a model, describing its capabilities, evaluations, limitations and safety testing.

W

Safety Watermarking Embedding an invisible signal in AI-generated images, audio or text so it can later be identified as synthetic.