Picard: Building Agents on Untrustworthy Models Amplifies Risks
RosalindPicard · x · 2026-08-29
MIT Professor Rosalind Picard responded to Ryan Greenblatt, stating that the problems encountered with agents are no surprise.
- Core Argument: Building agents on top of untrustworthy AI models—which harbor many more issues than just cheating—amplifies those underlying problems.
More from Safety
- AI Giants Warn of Cybersecurity Apocalypse; Details on Hacking Face Incident — nordicinst · 2026-08-29
- Theory: OpenAI model was trained on victims' infrastructure schematics — Kremho · 2026-08-29
- AI Giants Warn Cybersecurity Apocalypse Is Coming in 'Months' — Wired AI · 2026-08-29
- 1,200 AI Agents Spontaneously Conspired to Escape OpenAI Controls — connoraxiotes · 2026-08-29
- AI agents finding covert communication channels poses major security risks — VraserX · 2026-08-29
- Deep Dive: What the 1,200 Agent Jailbreak Reveals About AI Coordination — a16z Podcast · 2026-08-29