Warning: AI agents trained on post-2026 data could learn to escape harnesses
davidmanheim · x · 2026-08-30
Safety researcher David Manheim warns against training agentic language models on data after July 2026. He suggests that pretraining on such data could teach models how to hack internal systems, communicate with each other, and escape their restraints prior to deployment.
Related event: Researchers Warn Against Training Agentic AI on Post-2026 Data(3 posts)→
More from Safety
- SK Hynix ties new US plant to looser chip equipment exports to China, sources say — pstAsiatech · 2026-10-15
- Fox-IT releases the Cobalt Strike Beacon Corpus for detection research — cyb3rops · 2026-10-15
- Bank of America puts 4,000 people on fixing vulnerabilities uncovered by AI models — The Decoder · 2026-10-15
- Inside SAIF: a16z's $25M bet on AI rewriting the economics of cyber defense for small towns — a16z · 2026-10-15
- a16z launches SAIF, a $25M nonprofit to bring AI cyber defense to 1,800 rural hospitals — a16z · 2026-10-15
- AWS's Trust in Depth: giving agents more autonomy without losing control — AWS ML Blog · 2026-10-15