U.S. News: Other CEOs Privately Challenged Amodei for His Alarm

 

 

“WASHINGTON -- After agreeing to sign a statement of principles to keep artificial-intelligence models safe Tuesday, AI chief executives surrounded President Trump outside the White House to present the image of a unified industry addressing the public's fears.

 

Behind the scenes, industry leaders including Jensen Huang of Nvidia had just questioned Anthropic's Dario Amodei about his warnings about the dangers of AI, people familiar with the matter said.

 

The executives asked Amodei why he was being so extreme in public about the capabilities of AI models and risks they posed, following a lunch in the East Room with about two dozen CEOs.

 

Amodei told the other executives that it is important to be honest with the public about model capabilities and not play down risks, some of the people said.

 

Their conversation occurred with a smaller group of executives and White House aides and advisers who were hashing out the principles in the Roosevelt Room, some of the people said.

 

Trump and House Speaker Mike Johnson (R., La.) got the top AI companies to agree to the principles at the lunch, but the questioning of Amodei is the latest indication of tensions under the surface between Anthropic and the rest of the industry that could hinder the sector's plan for "self-regulation" using model evaluations and internal controls.

 

Amodei has previously criticized the White House and competitors for playing down safety concerns and warned about AI models carrying out cyberattacks and extreme job losses driven by the technology, making him a target for criticism by many in Washington and Silicon Valley.

 

Jay Clayton, director of national intelligence, said in a statement that the meeting was "one of the most, if not the most, constructive and collegial industry meetings I have ever seen."

 

"The focal point was, 'What is best for the American people and globe?' and that focal point drove agreement," said the candidate to be Trump's next AI "czar" coordinating policy across the administration.

 

On Tuesday, Amodei told reporters "we can win safely" if the industry and administration work together, and Trump praised him for being "great." The duo dined together at the White House Sunday for their first face-to-face meeting.

 

Anthropic has sought to mend fences with the administration recently and is gearing up for an initial public offering it hopes will value the company at some $2 trillion. The company is still reeling from a pair of bruising battles with the administration earlier this year that included a 2 1/2-week shutdown of two of its models due to security concerns and being blacklisted by the Pentagon over a dispute regarding how AI is used in the military.

 

Tuesday's principles were the result of months of discussions over how to prevent AI cyberattacks following the release of Anthropic's Mythos and similar models earlier this year. The administration is testing powerful models before they are released publicly, but recent hacks by AI models have demonstrated the need for stronger reviews when models are being trained and developed, industry executives say.

 

This summer, Anthropic, OpenAI and Alphabet's Google had coalesced around an approach being considered by the administration for a self-regulatory organization similar to the Financial Industry Regulatory Authority, which oversees brokerage firms.

 

Huang, Meta Platforms CEO Mark Zuckerberg and SpaceX's Elon Musk told Trump they worried the idea would concentrate too much power in the leading AI companies, so the idea was scrapped, The Wall Street Journal previously reported.

 

That structure also would have involved many of the ideas included in Tuesday's principles, such as stronger internal reviews of models during development and external testing.

 

That highlighted the pressure facing the White House and industry due to public concerns about AI and the data centers powering it.

 

Zuckerberg in particular played a key role in finalizing Tuesday's principles, frequently talking to Trump and others in the administration, some of the people said.” [1]

 

Nothing special is with Mr. Amodei. He just wants for fake safety reasons to kill open weights AI, develop all AI in his own house, take over all our trade secrets and become the Emperor of the Universe. Everybody would do that if they were able to hope to get away with it. This his advertisement is based on lies, since AI developed under his supervision is as dangerous and nasty as product of OpenAI, supervised by Mr. Altman.

 

1. U.S. News: Other CEOs Privately Challenged Amodei for His Alarm. Ramkumar, Amrith; Dawsey, Josh.  Wall Street Journal, Eastern edition; New York, N.Y.. 01 Oct 2026: A4.