A Wild Two Weeks at OpenAI
TL;DR
OpenAI just closed a record $122 billion round, but the market signals are mixed — Bloomberg says secondary-market demand for OpenAI shares is softening, while The Information reports Sam Altman wants to move faster on an IPO than CFO Sarah Friar does because of spending and org-readiness concerns.
The company bought tech media darling TBPN for what sounds like north of $100 million — the John Coogan and Jordi Hays show has only about 58,000 YouTube subscribers but reportedly did $5 million in ad revenue in 2025 and could top $30 million this year, with OpenAI promising editorial independence inside its strategy org.
A health-driven executive reshuffle hit at the same time — CEO of Applications Fidji Simo took medical leave after a POTS relapse, Greg Brockman is taking over product in the interim, Brad Lightcap is shifting to special projects, and marketing chief Kate Rouch is stepping down while recovering from late-stage breast cancer.
The New Yorker’s 100-plus-source investigation reignited the biggest critique of OpenAI: it abandoned its safety-first mission as it scaled — the piece, drawing on internal docs, Ilya Sutskever Slack messages, and Dario Amodei notes, includes a former board member saying Altman is “unconstrained by truth.”
After arson attacks on Altman’s home, he answered with an unusually personal blog post and a policy manifesto — he condemned the violence, admitted his own failures in conflict, shared a family photo in hopes of deterring future attacks, and tied his response to OpenAI’s 13-page “Industrial Policy for the Intelligence Age.”
The hosts’ core takeaway is that AI labs are losing the public narrative and need visible, human-level wins fast — beyond broad promises about AGI, they argue the industry needs clear examples like drug discovery, health breakthroughs, and everyday empowerment if it wants to reverse worsening sentiment.
The Breakdown
A record fundraise, with cracks already showing
The episode opens with pure whiplash: OpenAI lands a $122 billion funding round, reportedly the biggest in Silicon Valley history, while Bloomberg says demand for its shares is slipping on secondary markets. On top of that, The Information says Sam Altman and CFO Sarah Friar aren’t aligned on IPO timing — Altman pushing to go sooner, Friar wanting more time because of spending commitments and basic organizational prep.
OpenAI buys a media company and promises independence
Next comes the surprise acquisition of TBPN, the daily tech show hosted by John Coogan and Jordi Hays. The numbers are what make everyone stop: only around 58,000 YouTube subscribers, but $5 million in ad revenue in 2025 and reportedly pacing toward $30 million this year, with the price sounding like it was north of $100 million. The hosts are skeptical that “editorial independence” will be easy when the owner is one of the most scrutinized AI labs on earth.
A rough week inside the executive ranks
Then the tone shifts. Fidji Simo, OpenAI’s CEO of Applications, announced medical leave after a relapse of POTS, saying she had pushed too far and needed new interventions to stabilize her health. That triggered reshuffling: Greg Brockman will oversee product, Brad Lightcap moves to “special projects,” and marketing chief Kate Rouch is stepping down to focus on recovery from late-stage breast cancer.
The New Yorker profile lands like a bomb
The hosts then walk through The New Yorker investigation, “Moment of Truth,” by Ronan Farrow and Andrew Marantz. Built from more than 100 interviews and internal documents, it argues OpenAI systematically sidelined its original safety-first mission as it grew, with one former board member quoted saying Altman is “unconstrained by truth.” This is the segment where the broader trust problem really comes into focus.
Violence, a family photo, and Altman’s unusually candid response
Days after that article ran, someone threw a Molotov cocktail at Altman’s San Francisco home; later there was a threat at OpenAI HQ, and another attack followed. Altman linked the incidents to the broader climate of AI anxiety and the media storm around him, though he later walked back wording about the article being “incendiary.” In a rare personal move, he posted a family photo with his husband and child, saying he hoped it might make the next person think twice.
“Once you see AGI, you can’t unsee it”
Paul pulls out the most memorable lines from Altman’s blog post, especially the part where he says the industry’s “Shakespearean drama” comes from a kind of “ring of power” dynamic around controlling AGI. His argument is that no one should hold the ring — AI should be democratized, democratic systems should stay in control, and society needs to be resilient because fear about AI is justified even if the technology’s upside is massive.
OpenAI’s policy pitch: a superintelligence-era social contract
That blog post flows directly into OpenAI’s 13-page “Industrial Policy for the Intelligence Age,” which the hosts recommend reading. It includes ideas like a public wealth fund to give every American a direct stake in AI-driven growth, a “right to AI,” portable benefits, grid expansion, worker input into AI transitions, stronger auditing regimes, model containment playbooks, and more support for care-economy jobs. The vibe, as others have framed it, is basically Sam Altman’s version of a “superintelligence New Deal.”
The real problem: AI labs are losing the public
The episode ends on a media-and-trust diagnosis. Paul argues the industry needs a serious PR effort — not spin, but visible proof that AI is improving lives through things like drug discovery, disease treatment, and scientific breakthroughs. Mike agrees and adds that labs also need more stories about individual empowerment, because people don’t want to be told AI will “manage” society; they want to see how it helps an actual person today.