Debate erupts over Microsoft's 'dangerous' stance on model self-presentation
NinaPanickssery · x · 2026-09-18
After rgblong argued that Microsoft's takes on model self-presentation seem counterproductive and dangerous from a safety standpoint, Nina Panickssery pushed back: isn't this basically Roko's basilisk reasoning, and do you really have so little faith in aligning models to obedience that no lab should even try? The exchange continued into rgblong's specific concerns about embedding contradictory views of consciousness, goals, and self in models.
Related event: Microsoft AI CEO Suleyman Warns Anthropic May Be Training a Moral Agent(16 posts)→
More from AGI Musings
- Databases to distributed systems: every systems skill is merging into AI inference and hardware — Abhishekcur · 2026-11-02
- Limitless analogy: smarter people may gain far more from superintelligence — McDonaghMatthew · 2026-11-02
- Hunting for the most concrete "curing cancer with AI" roadmap of 2025 — iskander · 2026-11-02
- Avoiding AI brain rot: Miles Cranmer's lecture on learning with AI — _ScottCondron · 2026-11-02
- Yacine mocks "autonomous research" hype as Naam argues RSI loop needs 5-10x boost to run away — yacineMTB · 2026-11-02
- Vitalik's "post-AI cryptography": pseudonyms are dead, per-action anonymity is in — devanshmehta · 2026-11-02