Last week in a federal courtroom in Oakland, Sam Altman spent four hours on the witness stand explaining how OpenAI got from a nonprofit research lab to the world's most valuable private company. The trial in Musk v. Altman is now in its third week, with Elon Musk seeking up to $150 billion in damages over OpenAI's for-profit conversion. Yesterday, after less than two hours of deliberation, the jury ruled in Altman's favor, dismissing all of Musk's claims as untimely under the statute of limitations. Musk says he will appeal, arguing the verdict was a "calendar technicality." The whole proceeding was essentially an argument over a 2015 founders' agreement that no one bothered to write down.

The discovery process has produced a remarkable view into how decisions were made at one of the most consequential companies of the decade.

Greg Brockman's, OpenAI co-founder, private journal were entered into evidence over his lawyer's objections.

Internal Microsoft emails from 2018 surfaced where Satya Nadella admits, "Overall I can't tell what research they are doing."

In a 2018 late-night strategy meeting Altman described as "a long conversation of him [Musk] showing us memes on his phone."

Musk texted Brockman two days before trial: "By the end of this week, you and Sam will be the most hated men in America."

These are the ways where the foundational decisions of the AI era got made and they look a lot less like board strategy and a lot more like group chats with VERY large checks attached.

The same dynamic showed up last week, not in a court room, but in a ceremonial room in China. U.S. President Donald Trump traveled to Beijing aboard Air Force One with a powerful tech delegation that included Tim Cook (Apple), Jensen Huang (Nvidia), Elon Musk (Tesla and SpaceX), Sanjay Mehrotra (Micron), and Dina Powell McCormick (Meta), along with several Wall Street CEOs. The goal, among other things, was to meet face-to-face with Chinese President Xi Jinping to ease diplomatic tensions and increase economic partnerships.

The guest list on this trip was very intentional and shows the operating model for AI policy in 2026. There area a handful of executives in close proximity to power, making decisions that affect chip flows, model deployments, and international agreements.

Let's get into it.