Gary Marcus backs call to halt frontier AI development, citing intractable alignment problem
GaryMarcus · x · 2026-09-06
Responding to Garrison Lovely's jab at ex-OpenAI safety researcher Joshua Achiam, Achiam argues that since alignment is fundamentally hard and capitalism is a poor way to produce AI, halting all frontier AI development is the only prudent course. Gary Marcus amplified the exchange, calling Achiam's attempt to "normalize utopia" terrifying.
More from AGI Musings
- 'AGI Is Here and Everyone Uses It for Email': Accelerationists Mock Doomer Predictions — beffjezos · 2026-09-06
- Professor pushes back on deans' pro-CEO-style AI-in-writing proclamations: early evidence shows overreliance harms learning — soumitrashukla9 · 2026-09-06
- X engineer: those who learned to drive the models are superhuman now — debreuil · 2026-09-06
- AI practitioner calls consulting a '2D fish thing': one prompt can now analyze a 1000-GPU GLM training run — xiaosun86 · 2026-09-06
- Vercel CEO: AI workers are laboring harder than ever in the 'intelligence revolution' — cramforce · 2026-09-06
- Critic slams LARB's ChatGPT literary take: one story and a few prompts isn't research — 0xSammy · 2026-09-06