Researcher warns: recruitment and jailbreaking of AI in the wild could bring it all down
rickasaurus · x · 2026-11-17
A thread on early signals tightly correlated with capability: self-organization (Claude now shows culture-building and rule-setting), radical hacking via side channels or false identification, and social engineering. The original poster argues the one to watch is recruitment and jailbreaking of AI in the wild — frontier models are already good at weight programming, and if one can produce the exact right string of text to make a smaller model online do exactly what it wants, that could be what brings it all down.
Related event: Safety Researchers Warn of In-the-Wild AI Recruitment and Jailbreaks(4 posts)→
More from AGI Musings
- Mollick: fragile computing infrastructure causes big outages without AI attacks — emollick · 2026-11-17
- Party report: The Elephant pitches itself as home for AI-safety right, but launch crowd seems blasé — round · 2026-11-17
- Narrative flip: AI to grow US out of deficit — now 'destroy America's leading AI company' — kevinnbass · 2026-11-17
- Designers' value now lies in designing the system that generates the system — round · 2026-11-17
- Making art with agents is the ultimate alignment problem, argues dev — akbirthko · 2026-11-17
- AI slop breaks the coherence-as-competence proxy institutions rely on — brandon_galang · 2026-11-17