Ex-OpenAI researcher So8res: labs should collude on safety and dare DOJ to sue

trevposts · x · 2026-11-18

Former OpenAI researcher Nathan So8res (@So8res) argues in a discussion with @trevposts and Matt Levine that frontier labs don't need elaborate mechanisms — they should simply collaborate to avoid the literal apocalypse and dare the Justice Department to sue. He notes the resulting antitrust case would itself be extraordinarily valuable for convincing the world that the current AI race is horribly reckless, putting the safety-vs-competition tradeoff on public trial.

Original post →

More from AGI Musings

AGI Musings channel →