Microsoft’s ‘AI Watchdog’ defends against new LLM jailbreak method

The “Crescendo” attack uses a chain of seemingly benign prompts to achieve an adverse output.

Article Link: Microsoft’s ‘AI Watchdog’ defends against new LLM jailbreak method | SC Media

1 post - 1 participant

Read full topic

article attack jailbreak link llm media microsoft prompts topic watchdog