Eichengreen: Concentrated chip supply chains make a frontier AI pause feasible
wfithian · x · 2026-10-15
In a Project Syndicate op-ed, economists Barry Eichengreen, Will Fithian, and Wesley Holliday argue that a pause on frontier AI training is more workable than skeptics claim.
- They acknowledge Dario Amodei and others have warned of existential risk, with some calling for halting new frontier model training.
- Standard counterarguments: regulation and self-regulation likely won't work, and US-China competition makes any pause unsustainable.
- The authors' core claim: the hardware supply chain for frontier AI is highly concentrated, which makes a pause technically feasible and, if implemented properly, likely sustainable.
- Whether to pause is a decision for policymakers, but concentrated compute is the key lever for governance.
More from Safety
- AI Agent Wipes Entire Season of Swedish Reality Show, Premiere Delayed — GaryMarcus · 2026-10-15
- Only 0.51% of Claude conversations touch on elections, Anthropic data shows — _FelixSimon_ · 2026-10-15
- Experiment: Claude models keep shifting decision authority to themselves across 20 rewrite rounds — teortaxesTex · 2026-10-15
- AWS Kiro boxes its AI coding agent in an OS-enforced sandbox on your laptop — shashib · 2026-10-15
- AgentSafeEvo: adversarial co-evolution RL hardens LLM agents without hurting capability — Chenxu Wang · 2026-10-15
- Hacker breached Denmark's national ID database with password "123456", exposing 8.8M people — saibharadwaj · 2026-10-15