Trump invites Amodei to a private White House dinner
For two weeks, Anthropic's chief executive has been the administration's favourite example of AI doomerism. On Sunday evening, he is the guest.
President Trump plans to host Anthropic chief executive Dario Amodei at a private White House dinner on Sunday evening — the first one-on-one between the two, and, according to Axios, a sign that relations between Anthropic and the administration are thawing. Axios's Maria Curi and Marc Caputo, citing sources familiar with the plan, report that Trump extended the invitation personally, after Amodei was the conspicuous absence from last week's state dinner for Xi Jinping — the room where Sam Altman, Sundar Pichai and Jensen Huang were seated. The dinner, they write, "serves as an appetizer" for Tuesday's White House meeting between Trump, House Speaker Mike Johnson and the top AI chief executives. That is the date we flagged on Saturday, when the summit and the Amodei–Huang regulation split landed on the same guest list — Trump's AI week: America.gov goes live Tuesday with Huang and Musk.
Two things are true at once. Anthropic has been losing fights with this administration all month: on Friday a federal appeals court upheld the Pentagon's decision to designate the company a supply chain risk, on a 2-1 vote — Appeals court upholds Pentagon's Anthropic blacklist on a 2-1 vote — and the essay Amodei published on September 12 asking for outside evaluators inside his own company is the same argument Trump attacked as a foreign-policy mistake. And in an administration this attuned to the president's personal relationships, a dinner with him is worth more than a policy paper. Anthropic's staff will call this engagement; the plainer read is that it is court access, which is what a federally blacklisted vendor actually needs.
The caveat belongs next to the news, not after it: this is a single-source scoop. No second outlet has independently confirmed the dinner on its own reporting, the White House has said only that America will lead the world in "Super Intelligence" while protecting consumers, and no readout followed last week's state dinner either. A scheduled dinner is also not a reversal — the Pentagon designation still stands, and Tuesday's meeting will seat Amodei next to the Nvidia chief executive who says Anthropic is asking to be excused from laws the industry already has.
The same day, Beijing's answer to the West's doomer turn arrived: in China, warnings about existential AI risk are read as distinctly Western, or as a ploy to slow Chinese labs trying to overtake their American rivals. That is the framing in a New York Times dispatch from Lily Kuo, and Chinese-language coverage made the same argument the same day — the financial outlet 加美财经 published an analysis asking why China is not afraid of "AI doomsday talk." The real disagreement is about timing. Western labs argue that catastrophic risk justifies slowing down now, before capabilities outrun oversight; the Chinese read is that a country still catching up has far more to lose from a slowdown than from a speculative scenario. Worth noting that both positions are held by parties with an interest in the answer: the labs that warn loudest are also the labs asking to be trusted with regulating themselves.
What to watch: whether Sunday's dinner produces anything on the record, and whether Tuesday's meeting does.
If the loudest warnings about AI risk come from the company that just got a private dinner with the president, whose warnings should you weigh — and whose should you discount? Tell us in the comments.
