BSides talks featuring Claude

Anthropic's large language model shown to be susceptible to jailbreaking techniques across safety guardrails 4 talks featuring this tool from 4 chapters.