All In Podcast: AI Doom, OpenAI’s Math Win, and Nike’s Brand Slump

**SPEAKER_1** (0:01)
Power, panic, and prestige collide in this episode of All In. Chamath Palihapitiya, Jason Calacanis, David Sacks, and David Friedberg spend much of the conversation on one explosive question. Are AI doom warnings genuine, or a carefully amplified campaign? And the catalyst was Jacob Coxin, a former OpenAI and Anthropic researcher who quit and posted that, "The people building AI earnestly believe that it could kill us all by the end of the decade." The post went viral immediately, and Anthropic's alignment lead publicly agreed there was catastrophic risk potential. Right, within hours it was on cable news, on X, and dominating political conversations. But David Sacks pushed back hard. He said, "This is all just vibes. Show us the data, the report, or the leaked information." He argues the reaction looks less like whistleblowing and more like an organized push. That's where it gets interesting. Sacks pointed out how doomer groups and well-funded advocates quickly amplified the message. In his view, this wasn't spontaneous at all. And David Friedberg went even broader, saying, "We are in a hysteria phase of AI doomerism." He compared it to climate panic, COVID lockdown debates, and nuclear fear. The argument being that fear is natural when facing the unknown, but it can also become a mechanism for social control.
What do you think their real concern was here?
They believe the real target is open source. Friedberg said, "Open source is the game changer," because it lowers costs, spreads access, and prevents AI from being locked behind a few giant companies. Their warning is that a federal AI regulator could become a gatekeeper, protecting incumbents and blocking public models. That ties directly into the Anthropic IPO dilemma. If senior people at the company believe AI could be civilization-ending, how can investors price the business as a trillion-dollar platform? The panel suggests Anthropic faces a brutal choice: disavow the resignation as a doomer op, or agree with it and accept the consequences.
Building on that point, the conversation pivoted to OpenAI's reported math breakthrough on the Navier-Stokes problem. The panel treated it as proof that AI can massively compress human labor, not evidence of some mystical superintelligence. But that success raised practical warnings about data security. One speaker warned, "If you have sensitive, proprietary data, you cannot trust these LLMs." They're calling for sovereign infrastructure, stricter legal protections, and less blind trust in vendor promises.
Finally, they turned to Nike falling out of the S and P 100 after years of weak execution. The panel blames a drift away from mastery and product quality. "Nike was built on the back of legendary athletes who embodied mastery," one speaker said, and that's the standard they believe the company lost.
That's a fascinating parallel—whether it's AI companies or legacy brands, the message seems to be about maintaining authentic standards versus getting caught up in narratives that serve other agendas.