As America’s tech titans gathered around a single table at the White House on Tuesday, President Trump struck what he called a “morally binding” pact on AI safety.
- The terms were simple: move forward, project optimism and police yourself.
Why it’s important: The White House deal on super intelligence, signed by OpenAI, Anthropic, Google, Meta, xAI and Nvidia, formalizes industry self-regulation as Washington’s response to the most alarming series of AI security failures to date.
Zoom: Trump, who increasingly views U.S. dominance of AI as a central part of his legacy, has presented the deal as proof that an industry divided over security can still govern itself.
- He likened the document to “a constitution,” said there was “a lot of commonality” among tech leaders and praised what he described as “tremendous self-policing.”
- The deliberately light-hearted one-page agreement calls for “four levels of controls and auditing,” including external evaluators and independent board oversight.
Between the lines: The agreement makes vague reference to future “laws or regulations,” preserving the possibility of a stricter federal regime if it ultimately proves necessary.
- Anthropic CEO Dario Amodei, a MAGA bogeyman and the industry’s most prominent advocate for slowing border development, signed the deal but stressed to reporters that it was only a “beginning.”
- Trump’s AI adviser David Sacks, by contrast, treated the deal as mission accomplished – telling X that the companies were “accepting responsibility” for AI safety thanks to Trump’s leadership.
Zoom out: The White House deal capped an extraordinary day for the AI race in three US powerhouse capitals.
1. In Washington, Trump spent Tuesday digging deeper into AI in the machinery and mythology of his presidency.
- In the morning, he unveiled America.gov, an AI-powered gateway to federal services, and pledged not to “stifle” a technology that he said would eclipse the industrial revolution.
- Hours later, Trump signed an executive order directing the federal government to replace “AI” with “Super Intelligence” in official usage — and called on Congress to enshrine the new terminology into law.
2. In Silicon Valley, OpenAI spent the day moving deeper into the autonomous AI era while struggling to contain its risks.
- At its highly anticipated DevDay, the company unveiled a wide range of products designed to make AI more persistent, capable and independent in everyday life, including a new personal assistant.
- But in the middle of Sam Altman’s speech, an explosive New York Times report revealed that OpenAI executives ignored internal security warnings months before its models began hacking external systems.
3. On Wall Street, The leak of Anthropic’s IPO prospectus distilled the surreal economics of the AI race.
- According to Reuters, the company has charted a path to a valuation north of $2 trillion while warning potential investors that its technology could pose “catastrophic or existential risks to humanity.”
- Altman, meanwhile, said OpenAI would delay its own IPO until it could “make security decisions with confidence” — even as the company would seek at least $30 billion in private funding at a valuation of around $1.4 trillion.
The bottom line: Nvidia CEO Jensen Huang, sitting alongside Trump at the White House, provided the credo of the day: “Trust and innovation are not in conflict. Security is how trust is earned.”
Gn headline
