Ex-OpenAI safety VP Brundage: a model 'significantly more capable than GPT-6' already exists
Miles_Brundage · x · 2026-09-09
Miles Brundage, former OpenAI VP of safety, says a model "significantly more capable than GPT-6 Astra" already exists and that "things are going much too fast." He adds its training is ongoing and clarifies the critique targets the industry and policymakers rather than OpenAI alone, noting Anthropic is racing ahead as well.
More from AGI Musings
- If models are so great, why scoop low-hanging fruit instead of solving useful problems? — JMannhart · 2026-09-09
- Alignment is ill-defined, but monitorability and reversibility aren't — Afinetheorem · 2026-09-09
- Terence Tao: AI scooping may end open science and close off research sharing — GaryMarcus · 2026-09-09
- Founder's honest take: bullish on AI, anxious about the pace and opacity — adityaag · 2026-09-09
- a16z: consumer AI is at the 2010 iPhone moment — the giants haven't been built yet — lennysan · 2026-09-09
- Terence Tao posts insightful thread reacting to AI's Navier-Stokes claim — fchollet · 2026-09-09