Stanford Prof Calls for Coordinated Slowdown on Frontier Models Before Disaster Strikes
anshulkundaje · x · 2026-09-15
Stanford professor Anshul Kundaje argues frontier labs keep shipping half-baked models without guardrails while claiming they need time to add safety measures — and in a race, someone else ships the same flaws anyway, so a coordinated slowdown is the only fix.
Key points from the thread:
- Punitive laws only work after a catastrophic lesson; he favors heavy penalties even for recent relatively benign incidents.
- He'd welcome the 4–5 major developers agreeing to third-party oversight and vetting before release, with analogous community-run verification ideally applied to open models.
- A pushback frames his stance as self-inflicted and irresponsible, which he concedes — hence the call for coordinated action before a real disaster forces change.
Related event: Stanford professor calls for coordinated slowdown of frontier AI releases(3 posts)→
More from AGI Musings
- Critics Slam AI Firms for Hyping Danger While Disclosing Zero Technical Detail — 1a3orn · 2026-09-16
- Virologist vs. AI safety advocate clash over whether AI biorisk is real or a regulation dodge — anshulkundaje · 2026-09-16
- Investigation: EA Donors Paid Guardian $3M+; All Six in AI-Doom Article Funded by Same Three Foundations — mark_k · 2026-09-16
- Is the Brain a Computer? A Debate Over LLMs, Intelligence and Hidden Dualism — WillRinehart · 2026-09-16
- Max Levchin: Publicly discussing agent takeover strategies feeds the training data — JosephJacks_ · 2026-09-16
- Nobody Got Fired for AI Skepticism in 2023 — Does That Still Hold in 2026? — dfinke · 2026-09-15