David Deutsch updates views on AGI alignment and international coordination
danfaggella · x · 2026-08-19
Dan Faggella summarizes David Deutsch's updated views on AGI:
- Prosaic alignment works: AIs are likely to be trustworthy and aligned, leading to good outcomes.
- International alignment is less likely: Efforts like the UK AI Summit have been watered down in Paris, weakening coordination.
- Conflict risk: Reduced misalignment risk lowers the incentive for nations to coordinate. He predicts nations may restrict access to powerful models for citizens while using them for military/economic gains, leading to potential conflict.
More from AGI Musings
- Studying AGI reachability through formal logic and computability theory — suayptalha · 2026-08-19
- 90% of AI transformations stall on competing priorities, not tech, per BCG survey — alex_verem · 2026-08-19
- Mathematician's View: AI Partial Automation Isn't the End, Needs Guidance — aiamblichus · 2026-08-19
- AI might not replace SaaS, but could make it significantly more useful — Elegant_Map_7518 · 2026-08-19
- Paper Proposes Set Theory-Based AGI Architecture with Human-Level Efficiency — MikePFrank · 2026-08-19
- LLMs Are Now Superhuman Forecasters and Understand Human Behavior Better — paraschopra · 2026-08-19