They told the world that AI was dangerous. Now Trump’s allies are coming for them.

BERKELEY, Calif. — After artificial intelligence security advocates caught Washington’s attention with dire warnings about the technology’s power, the Trump administration and its allies adopted some of their suggestions, while sidelining the advocates themselves.

Battle lines have been drawn in recent days over part of an agreement signed last week by President Donald Trump and AI executives: who should audit AI companies’ systems to ensure their technology behaves as expected.

Subscribe to The Post Most newsletter for the biggest and most interesting stories from The Washington Post.

Leading AI companies — including OpenAI and Anthropic — have worked for years with a group of nonprofits concentrated in the Bay Area’s AI security community that grew out of the effective altruism movement. But Trump’s allies in the tech world say the job of auditing AI should go to America’s biggest consulting firms.

“In my opinion, it should be professional auditors, not these NGOs,” investor and former White House AI adviser David Sacks said last week on his All-In podcast. He said in an article on X last month that METR, an AI security nonprofit that worked with Anthropic, was too close to the company to be called independent. Shortly after, Anthropic announced it would partner with consulting giant Accenture to assemble a team of external reviewers.

At a conference here this weekend, members of AI security organizations grappled with the impact of what some see as a coordinated campaign to delegitimize their work and portray their community as bigoted and unserious. The event took place at an old hotel popular with AI security acolytes that also recently hosted an AI consciousness conference and sex workshop called “Slutcon.”

Trump officials and his allies have criticized the move for months, but the attacks intensified after public concern over AI risks forced the White House to respond. Sacks shared on X last month a New York Post front page that called METR employees, who helped OpenAI investigate its rogue AI agents, “THE WOKE WIZARDS OF AI.” (The Washington Post has a content partnership with OpenAI.)

Trump on Sunday named several figures who have rejected calls from advocates for a slowdown in the AI ​​industry to his Super Intelligence Force advisory board. Among them are Sacks and Emil Michael, the Pentagon’s chief technology officer, who said last month that calls for action to prevent extreme scenarios where AI would cause harm promoted “philosophies of extinction and death worship.”

“There has been a real rhetorical push in some corners of the White House and the broader Trump administration to get the message out about effective altruism and its potential dangers,” said a Trump administration official, speaking on the condition of anonymity to describe the private discussions.

A White House official said he supports organizations that evaluate the safety and security of how AI models are designed. The official pointed to Trump’s 2025 executive order targeting what the president called “woke AI.”

Anthropic declined to comment.

Nathan Leamer, director of the pro-industry nonprofit Build American AI, said he briefed congressional staff and other influential conservative figures in Washington about effective altruism, describing it as antithetical to his own Christian values.

“A year ago, no one in Hill’s office knew what effective altruism was,” Leamer said. Build American AI is linked to Leading the Future, a political campaign group funded by venture capital firm Andreessen Horowitz and OpenAI Chairman Greg Brockman, a major midterm Republican donor.

Brad Carson, co-founder of Americans for Responsible Innovation, which says it focuses on AI policy in the public interest, said Trump allies are trying to discredit anyone who advocates for meaningful regulation of AI, an idea that now has significant momentum.

“In Washington, at least, and perhaps in the broader culture, we have won this fight. The mood is with us,” said Carson, who also runs a political advocacy organization funded in part by Anthropic. “People are very concerned. Congress wants to regulate.”

But leaders of AI assessment startups that aren’t from the AI ​​security movement said some Washington Republicans are wary of the label.

“It almost seemed like a new version of McCarthyism,” Rayan Krishnan, CEO of Andreessen Horowitz-backed startup Vals AI, said after a trip to Washington last month to pitch his startup to the public sector.

Judd Rosenblatt, CEO of AE Studio, who describes himself as a conservative small business owner, said several people in the Trump administration told his team they saw the need for third parties to evaluate AI tools, but were wary of effective altruism, he said.

Civil society groups that have been auditing AI systems long before ChatGPT’s debut warn that there is a risk that an opportunity for oversight of opaque AI giants will be squandered.

“We’re now trying to figure out who will actually participate in decision-making,” said Miranda Bogen, chief technologist at the Center for Democracy & Technology and director of the advocacy group’s AI Governance Lab, one of several groups eager to contribute expertise in standard-setting.

The agreement signed by AI executives last week, which Trump called “morally binding,” was frequently mentioned on stage and in meetings between sessions at the Berkeley conference last weekend, where attendees endured a heatwave with fans and ice cream.

During a Sunday panel discussion on what the AI ​​auditor ecosystem could look like, one speaker said they were working to build a broad coalition of experts, including people working on bias and discrimination as well as blue-chip companies like Accenture and Deloitte. Panelists also discussed other ways the community could prove its legitimacy. Another suggested two-month boot camps to train a talent pipeline of qualified auditors.

Adam Gleave, CEO of FAR.AI, a nonprofit research institute, said in an interview that excluding the AI ​​security community would exclude people with significant expertise at this time.

“When it comes to looking at the transcripts of millions of agents and trying to piece together what they did and how they were interacting, I don’t think there are many organizations in the world that have that expertise,” Gleave said. “The reality is that anyone you hire with that experience will have some connection to the industry.”

On his podcast, Sacks and his co-hosts, who remain close to the White House, have made it clear that they prefer to leave the job to established giants.

“You’re going to see EY announce something soon,” said investor Chamath Palihapitiya, who said earlier in the episode that his startup 8090 was building the global accounting firm formerly known as Ernst & Young “a complete auditing infrastructure for super intelligence,” and that 8090 now plans to “roll it out to everyone.”

EY said in a statement that it was working with 8090 to “develop new capabilities that bring greater visibility, traceability and human accountability to consequential AI-driven decisions and workflows.”

Related content