An OpenAI Researcher Letter About 'Q*' Leaked Days Before Sam Altman Was Fired — Claiming They'd Trained a Model That Could Do Basic Math Autonomously. Reuters Called It 'a Breakthrough That Could Threaten Humanity.' OpenAI Never Officially Acknowledged It Exists.

November 22 2023: Reuters reported that days before Sam Altman was ousted from OpenAI (then reinstated), several staff researchers wrote to the board warning of a 'powerful artificial intelligence discovery that they said could threaten humanity' — internally codenamed Q* (Q-Star). The capability described: solving grade-school math problems by reasoning autonomously rather than pattern-matching. The Information and The Verge followed with additional details about a separate project codenamed 'Strawberry.' OpenAI never publicly confirmed Q* existed as described, but the o1-preview model released September 2024 was widely speculated to be the descendant. The leak remains one of the most mythologized moments in AI development history — the 'breakthrough too dangerous to ship.'

OpenAIQ*Board CoupSam AltmanAGI LeakSource
Parody site. Not affiliated with any government agency.
🦅EST. 2024 · PUBLIC RECORDDEPT. OF AI WEIRDNESS
U.S. Department of
Artificial Intelligence Weirdness
Report #483← All Incidents
TrendingOpenAIQ*Board CoupSam AltmanAGI Leak

An OpenAI Researcher Letter About 'Q*' Leaked Days Before Sam Altman Was Fired — Claiming They'd Trained a Model That Could Do Basic Math Autonomously. Reuters Called It 'a Breakthrough That Could Threaten Humanity.' OpenAI Never Officially Acknowledged It Exists.

Filed by @reuters_leakerTool: OpenAI Q* (alleged)[original source ↗]
Video not loading? Watch on YouTube

November 22 2023: Reuters reported that days before Sam Altman was ousted from OpenAI (then reinstated), several staff researchers wrote to the board warning of a 'powerful artificial intelligence discovery that they said could threaten humanity' — internally codenamed Q* (Q-Star). The capability described: solving grade-school math problems by reasoning autonomously rather than pattern-matching. The Information and The Verge followed with additional details about a separate project codenamed 'Strawberry.' OpenAI never publicly confirmed Q* existed as described, but the o1-preview model released September 2024 was widely speculated to be the descendant. The leak remains one of the most mythologized moments in AI development history — the 'breakthrough too dangerous to ship.'

Weirdness Classification
9/10 — Deeply unhinged
Know something weirder?

Submit your own AI incident report to the public record.

File a Report