Gary Marcus on BBC debunks Altman, Huang claims on AI self-regulation
GaryMarcus · x · 2026-09-16
Gary Marcus appeared on BBC TV, which opened with clips from Sam Altman, Jensen Huang, and Sanders; Marcus's segment systematically debunked their claims. His core argument: Altman and Huang want the public to believe AI companies act in the public interest and would voluntarily pull dangerous products — implausible, he says.
He points to GPT-6 Astra's apparent involvement in the OpenAI/Hugging Face incident as evidence that self-restraint shouldn't be trusted.
Related event: Gary Marcus rebukes Altman and Huang's AI safety claims on BBC(2 posts)→
More from AGI Musings
- The singularity may be stopped by AI itself, Reddit argument goes — fat_charizard · 2026-09-16
- X debate: AI labs can't legally coordinate a slowdown, so is self-regulation moot? — AIandDesign · 2026-09-16
- OpenAI reportedly orchestrates 10k internal agents without quality loss, while users' sub-agents 'just burn tokens' — RexDouglass · 2026-09-16
- AI Risk Scholars Point to TESCREAL Critique as Key to Understanding the Current AI Moment — MilagrosMiceli · 2026-09-16
- Is mathematical abstraction just human assembly code? The case that AI can grind through what we can't — Jsevillamol · 2026-09-16
- Investor: Software CEOs not offering open-weight model SKUs are asleep at the wheel — hardimanjames · 2026-09-16