EpochLex entry
AI Red Team
A-I RED teem
A group of internal or external testers who adversarially probe an AI model or system to uncover harmful capabilities, misuse pathways, security weaknesses, or failures in safeguards.
EpochLex entry
A-I RED teem
A group of internal or external testers who adversarially probe an AI model or system to uncover harmful capabilities, misuse pathways, security weaknesses, or failures in safeguards.