The White House has asked a group of tech companies to answer a set of questions this week about how to ward off digital attacks that frontier AI tools could soon enable, according to four people with knowledge of discussions between the administration and the tech sector.
The questions, from the White House’s Office of the National Cyber Director, focus on how specific sectors in the tech and cybersecurity industries can work with the White House to boost their defenses with AI, these people said. Companies have been asked to respond to them by Friday.
Some of the questions were discussed during a Tuesday afternoon meeting between White House cyber officials and approximately 30 industry representatives, said one of the people, who, like others in this report, was granted anonymity to share details of ongoing, highly confidential discussions.
The request for additional, detailed information from these companies reflects the intensifying focus in Washington on the evolving threat that hyper-advanced AI tools may pose to national security and digital infrastructure.
The Tuesday meeting was partly prompted by concerns about Anthropic’s newest AI model, Claude Mythos, and its ability to unearth hidden software flaws and outpace the world’s top hackers at certain tasks. The company has, for now, restricted access to a small group of security researchers and tech companies through what it has dubbed Project Glasswing.
Anthropic CEO Dario Amodei recently visited the White House to discuss Mythos with administration officials, including Treasury Secretary Scott Bessent, chief of staff Susie Wiles and National Cyber Director Sean Cairncross.
One list of questions sent by the White House to some tech and cyber firms, obtained by POLITICO, covers a range of technical and policy considerations, including which widely used coding projects should be prioritized and more basic questions about how the public and private sectors can work together on initiatives such as Project Glasswing. One question simply asks: “What is the most effective role for the government?”
The four people said some industry representatives were confused by the questions they received, several of which were seen as vague.
Two of those people said other questions addressed internal security practices some of these representatives did not feel comfortable disclosing to the government without a clear justification.
The questions include what systems they have used AI to test thus far and what their “scanning and remediation priorities are,” according to the list obtained by POLITICO.
The White House is also weighing executive action on AI, which two of the people noted was discussed during Tuesday’s meeting. Axios first reported on the possible executive action.
One U.S. official familiar with the discussion, who was not at the Tuesday meeting, said the draft executive order has undergone interagency review at the deputies’ level, though some resistance to the plan remains.
Spokespeople for ONCD and the White House did not respond to requests for comment.
The White House has been taking steps to defuse a monthslong legal battle with Anthropic over the company’s efforts to set ethical limits on government use of AI — a fight that led President Donald Trump in February to ban all federal agencies from using the AI company’s software. Since then, growing awareness of Mythos’ cyber prowess — as well as concerns that unauthorized users might be commandeering technology — has agencies clamoring for access to the tool.
Competitors, including OpenAI, have also recently begun testing their own advanced cybersecurity-focused AI models, creating additional pressure on the government to find a solution.
ONCD is seeking guidance on how to prioritize the avalanche of bugs this wave of advanced AI tools is likely to uncover, according to the list of questions obtained by POLITICO. This includes how to deploy software fixes to critical infrastructure operators without tipping off attackers and how best to share that information with other companies and the government.
POLITICO previously reported that Anthropic and OpenAI were among the companies that sent representatives to the meeting. Those at the meeting were asked to sign non-disclosure agreements, according to two of the people.
Lobbyists and policy officials recently told POLITICO that Mythos has led a faction within the White House to call for a détente with Anthropic. In March, the Defense Department formally designated the company a supply chain risk, a move that could eventually force companies contracting with the military to cut ties with Anthropic. The company is suing the administration in D.C. and California federal courts to challenge that label.
Multiple federal agencies in the U.S., as well as officials in allied nations, have requested briefings from Anthropic on Mythos’ hacking capabilities.
Things may be turning a corner. Trump said during a CNBC appearance last week that Anthropic’s executives are “shaping up.” He added, “They’re very smart. … I like high-IQ people, and they definitely have high IQs.”
