Leaked memos detail why OpenAI's board tried to fire Sam Altman

Sam Altman's 2023 ouster at OpenAI wasn't just boardroom drama—it was a coordinated, terrified coup attempt backed by a 70-page dossier compiled by co-founder Ilya Sutskever.

According to an extensive investigation published by The New Yorker based on over 100 interviews and internal records, Sutskever was so frightened of company surveillance that he used a personal phone to take photos of Slack messages and HR documents, sending them as disappearing files. The top item on his list of Altman's behavioral patterns was blunt: lying.

The dossier and the 'government-in-exile'

Sutskever had officiated president Greg Brockman's wedding in 2019, but as OpenAI neared human-level AI, he concluded that Altman was simply too untrustworthy to have his finger on the button. The memos alleged Altman repeatedly deceived the board about safety protocols and pitted executives against each other.

Altman learned he was fired while watching Formula 1 in Las Vegas. He immediately flew to his $27 million San Francisco mansion, establishing what he called a "government-in-exile."

Crisis PR veteran Chris Lehane, Airbnb co-founder Brian Chesky, and investor Ron Conway set up a literal war room. Lawyers camped in Altman's bedroom-adjacent office, and executives packed the hallways. While Altman poured Negronis every evening at 6 p.m. urging people to chill, phone logs showed him taking calls for over 12 hours a day.

Capital always wins

The board never stood a chance against Silicon Valley's economic machine. Microsoft was blindsided, with CEO Satya Nadella scrambling to figure out if Altman had committed embezzlement or sexual misconduct.

Venture firm Thrive was finalizing a deal valuing OpenAI at $86 billion, promising employees massive equity payouts. Thrive immediately froze the funding, signaling the cash only flowed if Altman returned. With Microsoft offering to hire the entire team and a massive staff revolt underway, the board folded. Altman was reinstated within five days, an episode staff now call "the Blip."

Why it matters

This investigation cuts through the mythology of OpenAI's idealistic nonprofit origins. It reveals a brutal reality: when civilizational safety clashes with multi-billion-dollar employee equity, capital wins every single time.

For enterprise customers and developers, the lesson is operational risk. Relying entirely on OpenAI means tethering your infrastructure to a leader whose closest scientific collaborators accused him of deception and an unchecked will to power. Diversifying toward open-source models and multi-provider pipelines is no longer just a technical precaution—it is governance insurance.

Altman's handling of the crisis was a masterclass in political hardball, but we think it leaves an enduring stain on OpenAI's safety credibility. Replacing independent board members with hand-picked allies while steering toward a potential trillion-dollar IPO proves that OpenAI is now just another aggressive Big Tech giant, regardless of its original mission.