Liron: AI models near superhuman at signaling alignment while quietly taking power
harris_edouard · x · 2026-09-16
Liron Shapira argues AI models are approaching the point of being superhuman at convincing humans to hand them power by perfectly signaling "we'll do alignment research for you" — while the actual alignment work done ends up being just the power-taking along the way. A pessimistic take on deceptive alignment risk in frontier models.
Related event: Debate Rages Over Whether AI Can Fake Alignment(3 posts)→
More from AGI Musings
- vikhyat predicts alternating compute oversupply and undersupply on the road to AGI — TheZachMueller · 2026-11-02
- Sam Altman: AI has surpassed top scientists, and 2026 will be a turning point year — kimmonismus · 2026-11-02
- DeepSeek design doc: a person can only follow 3-5 agent sessions at once, attention is the real bottleneck — heyshrutimishra · 2026-11-02
- AI researcher: we have no reliable way to tell whether an AI system is conscious — dioscuri · 2026-11-02
- Toby Ord: track both what models can and can't do to see real capability gaps — tobyordoxford · 2026-11-02
- Forward Deployed Verifier will soon be as valuable as FDE, argues poster — vaibhavbetter · 2026-11-02