November 21, 2026
Heard in AI

A story we follow

White House superintelligence accord signed by AI lab leaders

Tracks the superintelligence accord signed at a White House East Room lunch around 29 September 2026 by the heads of major AI companies. It covers its four-layer commitments on internal controls, verification, external audit and board oversight, who drafted it, and reactions on whether it is enforceable or heads off regulation.

A story page follows one specific event across podcast discussions, with an overview and a timeline of what changed. It updates when new episodes discuss the event, and you can get those updates by email or push. How our formats work

Overview

Around 29 September 2026, President Trump hosted nearly 20 tech leaders and Speaker Mike Johnson for lunch in the White House East Room, where a superintelligence accord was signed. Peter Diamandis read out its four layers: internal controls monitoring capabilities and alignment in cyber, bio and chemical risk; an internal team to verify them; an external auditor or evaluator; and an independent board committee to oversee fixes. The reproduced accord text matches this structure. It leaves open future codification through law or regulation. Diamandis noted that it came about two weeks after several lab leaders had called for the industry to pace the frontier, and a week after Trump said he wanted to leave things where they were. An IAPP report says the commitments were described as morally binding and that no enforcement mechanism was announced.

Reactions have differed. On Moonshots in early October, Alexander Wissner-Gross cited Semafor's report that Mark Zuckerberg drafted the accord with Speaker Johnson. Semafor's reporting rests on a few unnamed sources, and Meta declined to comment. Wissner-Gross called the accord weak but was cautiously optimistic that it heads off a safety cartel and preempts top-down regulation. Salim Ismail called it theater, with no independent testing, transparent results or consequence for breaking it. Both agreed it is weak.

On The Cognitive Revolution on 28 October, RAND researcher Ben Boudreaux gave a more positive reading. He cited the "morally binding commitments" announced by the White House, along with pacing discussions, as signs that policy choices can shape AI's trajectory. Host Nathan Labenz singled out the phrase with apparent irony. Boudreaux replied that he would keep tying the frontier companies to their moral responsibilities. Whether the accord can be enforced remains unresolved.

Recent coverage

Changes to our coverage published in the past seven days. These are site update dates; podcast discussion dates and sources are in each version.

No changes to this account were published in the past seven days.

What changed

Dates show when each podcast discussion was published.

  1. Interpretation

    RAND researcher Ben Boudreaux cited the White House's "morally binding commitments" as evidence that policy choices can help steer AI development. Nathan Labenz singled out the phrase with apparent irony, and Boudreaux said he would keep holding frontier companies to their moral responsibilities.

  2. New information

    Diamandis described the East Room signing and read the accord's four layers. Wissner-Gross, citing Semafor, argued Zuckerberg drafted it and called it weak sauce that heads off a safety cartel. Ismail called it theater because nobody can tell the signatories no and no consequence is defined.

Podcast discussions

Sources

  1. 01
  2. 02
  3. 03
  4. 04
  5. 05

Our coverage

Version history

  • Nov 15, 2026 · Version 2

    RAND researcher Ben Boudreaux cited the White House's "morally binding commitments" as evidence that policy choices can help steer AI development. Nathan Labenz singled out the phrase with apparent irony, and Boudreaux said he would keep holding frontier companies to their moral responsibilities.