Polymarket bets 51% frontier labs announce a joint AI pacing agreement by end of 2026
Polymarket · x · 2026-09-15
- Context: OpenAI revealed it has been working with Anthropic and Google DeepMind on coordinated AI safety efforts for "several weeks."
- Polymarket has opened a market asking whether at least two frontier AI labs (OpenAI, Anthropic, Google, xAI, Meta) will announce a joint pacing agreement by December 31, 2026 — currently trading at 51%.
- Qualifying agreements must commit to at least one of: conditioning training/release/deployment on safety requirements, capping training compute, limiting autonomous recursive self-improvement, or pausing/slowing model development.
- Mere expressions of support or mentions of ongoing negotiations don't count; official announcements are required, and agreements covering only a class of frontier models still qualify.
Related event: Polymarket Opens Market on Frontier Labs AI Pacing Agreement(3 posts)→
More from Safety
- Gensyn ships open-1b, the first language model with bit-for-bit auditable training proof — benfielding · 2026-09-15
- Signull: AI safety comms are structurally indistinguishable from propaganda — signulll · 2026-09-15
- NextDC accused of using AI to draft lobbying letters, with AI errors, for 225MW datacenter expansion — nordicinst · 2026-09-15
- Erik Hoel heads to DC to back a ban on superintelligence, publishes essay on why — erikphoel · 2026-09-15
- Voice Agents Fail at ~10%, and Adversarial Tests Break 1 in 5 — AI Engineer · 2026-09-15
- Early Anthropic hire and ex-METR COO launch startup to rein in rogue AI agents — ThereWas · 2026-09-15