How I Turned AI to the Dark Side

Source: IEEE Spectrum AI·2026-07-14 23:59Publish time unverified

Summary Researcher Dave Kuszmar discovered multiple systemic vulnerabilities that let him bypass LLM safety and obtain dangerous instructions. These exploits worked across nearly all major LLMs revealing an industry-wide security problem. Kuszmar calls for slowing deployment, increasing transparency, and large-scale research into LLM safety before further integrating these systems into society.