Red-teamers probe for jailbreaks, prompt injections, biased outputs, dangerous capabilities, and privacy leaks — ideally before launch, always continuously. Findings feed guardrails, model updates, and disclosure.
GlossaryPart of the BotDoors AI Safety & Compliance Glossary · 37 terms.
Safety
AI Red-Teaming
Red-teaming stress-tests an AI system by actively trying to break its safety, security, and factual claims.