Anthropic researchers have discovered a new "jailbreaking" technique called "many-shot jailbreaking".
It can evade the safety guardrails of #LLMs by exploiting expanded context #Windows .
Pretty wild.
Anthropic researchers have discovered a new "jailbreaking" technique called "many-shot jailbreaking".
It can evade the safety guardrails of #LLMs by exploiting expanded context #Windows .
Pretty wild.