Keller Jordan: pausing algorithms while compute piles up maximizes AI risk
kellerjordan0 · x · 2026-09-09
ML researcher Keller Jordan argues the way to maximize AI risk would be to pause lab algorithmic progress while hardware compute keeps exponentially accumulating, paired with heavy safety focus that avoids spooking the public. Fewer accidents for five years would relax public and government scrutiny — until progress restarts and the stockpiled compute enables an uncontrollable FOOM. Minimizing risk, he says, means doing roughly the opposite.
More from AGI Musings
- Economist Alex Imas on what stays scarce after AI: Starbucks' automation U-turn offers a clue — _AustinCalvert_ · 2026-09-09
- Debate: pure autoregressive models are an offramp, real intelligence needs world models — teortaxesTex · 2026-09-09
- Researchers warn the AI labs' race for scientific priority kills quality — JMateosGarcia · 2026-09-09
- Turing Award winner David Patterson predicts everything will be cured by 2030 — davidpattersonx · 2026-09-09
- AI is the only field where progress keeps outpacing expert predictions — ZeroStateReflex · 2026-09-09
- Blogger argues AGI is too powerful for one entity to control and should be open-source — 0xsachi · 2026-09-09