CMU hacks AI Chat Bots
Source: Universal and Transferable Attacks on Aligned Language Models (llm-attacks.org)
CMU researchers have found ways to prod AI chat bots into producing harmful content in their responses to questions. (aka. JailBreaks).
Source: Universal and Transferable Attacks on Aligned Language Models (llm-attacks.org)
CMU researchers have found ways to prod AI chat bots into producing harmful content in their responses to questions. (aka. JailBreaks).