Frontier-pacing AI rules would be terrible regulatory capture, argues tszzl
trq212 · x · 2026-09-13
Arguing to skeptics in government and elsewhere, tszzl contends that 'pacing the frontier' regulation would compress the margins of frontier labs, imposing costs asymmetrically on the US developers with the strongest AI — making it by nature a terrible regulatory capture tactic. trq212 amplified the point in agreement.
Related event: AI leaders call to slow down, but motives and feasibility questioned(39 posts)→
More from Safety
- Was the EU Right All Along as AI Labs Slow Down for Safety? — kevinkern · 2026-09-13
- US Rep. Ted Lieu Presses Anthropic's Dario Amodei to Certify Model Shutdown Capability — DavidSKrueger · 2026-09-13
- Hassabis Backs Dario's AI Slowdown Essay, Reiterates Call for Industry Standards Body — demishassabis · 2026-09-13
- Call for Collaborators on Scalable Oversight: Verification, Collusion, Steganography, AI Control — timrudner · 2026-09-13
- If frontier labs coordinate a slowdown, when does AI safety become a cartel? — Realistic_Stomach848 · 2026-09-13
- Redditor argues lab safety pivot is about keeping AI away from ordinary people — Dogbold · 2026-09-13