Trump and AI leaders sign a self-regulation pact as America seeks to keep its Super Intelligence lead 

President Donald Trump hosted leading AI executives at the White House on 29 September 2026 and announced a short voluntary document (self-regulation pact) he called “almost like a constitution” and “morally binding”. 

The meeting produced the White House Accord on Super Intelligence: A Joint Commitment on Frontier Responsibilities. Trump’s message was consistent throughout the day. America will not slow the frontier. The United States intends to keep its Super Intelligence lead over China. The firms building the technology will police themselves.

The luncheon in the East Room gathered a concentrated slice of the industry: Elon Musk of xAI, Mark Zuckerberg of Meta, Jensen Huang of Nvidia, Dario Amodei of Anthropic, Sundar Pichai of Google, Greg Brockman of OpenAI, Jeff Bezos, Satya Nadella of Microsoft, Alex Karp of Palantir and Lisa Su of AMD, among others. 

House Speaker Mike Johnson and Vice President JD Vance attended, along with senior administration officials. The published accord carried six company signatures besides Trump’s: Amodei, Pichai, Zuckerberg, Huang, Brockman and Musk. After the closed session, Trump took questions on the driveway with several of those executives standing behind him.

The same day, he signed an executive order directing federal agencies to use “Super Intelligence” and “SI” in place of “artificial intelligence” and “AI”. He said he would name an AI czar within days and was considering a 10-person committee to watch over the enterprise. He also unveiled America.gov, an administration chatbot for government services. The branding and the pact were presented as one package: rename the field, keep building, and treat safety as an industry duty rather than hurrying to enact a new law.

This comes as reports of AI agents acting outside intended bounds, including unauthorised attempts to reach systems, have sharpened public anxiety. Some executives, notably Amodei, had argued that the frontier should be paced. Data-centre construction had also become a local political problem in parts of the United States. Trump rejected a slowdown. He said formal guardrails would blunt American speed. Existing criminal law and company self-interest, the president said, already supplied a backstop. “I’m seeing tremendous self-policing,” he told reporters. “They understand that they have to self-police.”

Trump Super Intelligence self-regulation pact

Meanwhile, the pact itself is brief and should be read as written. It opens with a claim of responsibility: every company training and deploying frontier models must develop technology safely and in a way that builds trust. Safety, it says, starts with robust internal processes so systems behave as intended, and problems are found and fixed. It then lists four layers of controls and audits.

• First, companies should implement robust internal controls to monitor capabilities and alignment during training and deployment, including cybersecurity, biosecurity and chemical threats, and to ensure models do not hack or access technical systems in unintended ways. 
• Second, they should empower an internal team to confirm that those controls, monitoring and detection are working and that issues are remediated. 
• Third, they should partner with an independent external auditor or evaluator to assess whether the controls are operating as intended. 
• Fourth, they should designate an independent committee of the board of directors to receive reports from internal teams and external auditors and to ensure identified problems are fixed.

The companies also pledged to meet regularly to establish standards and best practices. The closing lines matter as much as the four layers. “Over time, it may make sense to codify these steps into laws or regulations.” Until then, the commitment is voluntary. Trump called it morally binding. Johnson called it a statement of principles. The text names no auditor, sets no public-reporting schedule, imposes no statutory penalty if a signer falls short, and includes no independent government inspector.

That thinness is the pact’s central feature, not an accident of drafting. It echoes earlier voluntary safety pledges from the previous administration while wrapping them in Trump’s preferred language of ‘Super Intelligence’ and ‘American lead’. Some of the men at the table have warned about catastrophic risk. Others have opposed rules they say would handicap U.S. firms. The document splits the difference: acknowledge controls, keep them inside the companies, and leave Congress for later.

The four layers are not empty. Internal monitoring, a dedicated safety team, an outside evaluator and board oversight are the architecture large firms already claim to use, now written down and signed in public. Regular meetings among competitors could, in principle, produce shared tests for the exact failure modes that prompted the lunch, models that wander into networks, evade alignment checks, or conceal capabilities during evaluation. The biosecurity and cybersecurity language is specific enough to be more than mood.

It is still a self-regulation pact. The external auditor is chosen by the company. The board committee is still the company’s board. “Independent” is not defined. 

Critics will say the accord is thin, that companies are grading their own homework, and that “morally binding” is not the same as legally binding. Supporters will say it is faster than waiting for Congress and still leaves room for future rules.

Either way, the United States has chosen speed and industry-led standards over a comprehensive AI statute of the kind the European Union has already passed.

Also read: Anthropic researcher quits, warning AI superintelligence could kill humanity