Scoop: White House workshops plan to bring back Anthropic

April 29, 2026

```json { "title": "White House Plans to Reverse Anthropic AI Ban With Executive Action", "metaDescription": "The White House is developing guidance and a draft executive order to let federal agencies bypass Anthropic's supply chain risk designation and adopt Claude Mythos.", "content": "<h2>White House Moves to Reverse Anthropic AI Ban, Drafting Executive Action to Allow Federal Agencies Access to Claude Mythos</h2>\n\nThe White House is developing guidance — including a draft executive action — that would allow federal agencies to work around Anthropic's supply chain risk designation and onboard new AI models, including the company's most powerful yet, Claude Mythos, according to a report published April 29, 2026 by Axios. The development represents a dramatic potential reversal for an administration that, just two months ago, directed all federal agencies to immediately cease using Anthropic's technology and labeled the company a national security threat.\n\n<h2>From Blacklist to Back-Channel: How the White House Shift Unfolded</h2>\n\nThe conflict between the Trump administration and Anthropic began in earnest on February 27, 2026, when President Trump directed all federal agencies to stop using Anthropic's AI products. Defense Secretary Pete Hegseth simultaneously announced a pending supply chain risk designation — a label that, according to legal analysis by Mayer Brown, had never previously been applied to an American company. The designation was formalized in letters dated March 3, 2026, issued under two legal authorities: 10 U.S.C. § 3252 and the Federal Acquisition Supply Chain Security Act of 2018.\n\nAt the center of the dispute was Anthropic's refusal to waive contractual restrictions on the use of its Claude AI for mass domestic surveillance and fully autonomous weapons systems — conditions the Pentagon had demanded as part of a renegotiation of a two-year, $200 million prototype agreement originally awarded to Anthropic in July 2025 by the Department of War's Chief Digital and Artificial Intelligence Office. That contract had made Anthropic the first AI model provider approved for use on classified government networks. In late February, OpenAI moved to claim that Pentagon contract after Anthropic was pushed out.\n\nAnthropic responded with lawsuits filed in two federal courts on March 9, 2026. The legal campaign produced a split outcome. U.S. District Judge Rita F. Lin in the Northern District of California granted Anthropic a preliminary injunction blocking the administration from enforcing its ban on Claude for non-DOD civilian federal agencies. However, on April 8, 2026, the D.C. Circuit Court of Appeals denied Anthropic's request to temporarily block the Pentagon's supply chain risk designation. In a unanimous decision, the appeals court stated that granting the stay "would force the United States military to prolong its dealings with an unwanted vendor of critical AI services in the middle of a significant ongoing military conflict."\n\nFollowing Judge Lin's injunction, the General Services Administration reversed course. On April 3, 2026, the GSA published a statement confirming it had withdrawn its removal announcement and restored Anthropic's offerings for federal buyers.\n\n<h2>Claude Mythos and Project Glasswing Changed the Calculus</h2>\n\nThe political dynamics shifted significantly on April 7, 2026, when Anthropic released Claude Mythos Preview — described as a new class of AI model built for cybersecurity, autonomous coding, and long-running agents. The release was gated, offered first to a controlled group of partners through a new initiative called Project Glasswing. According to TechCrunch, 12 named partner organizations and 40 total organizations received early access, including Amazon, Apple, Broadcom, Cisco, CrowdStrike, the Linux Foundation, Microsoft, and Palo Alto Networks.\n\nAnthropic committed up to $100 million in usage credits for Mythos Preview across Project Glasswing efforts, along with $4 million in direct donations to open-source security organizations. In internal testing, Mythos identified thousands of zero-day vulnerabilities across every major operating system and web browser, according to Anthropic's Project Glasswing page and reporting by CSO Online. A small group of unauthorized users accessed Mythos on the same day Anthropic announced its limited release to partner companies, according to Bloomberg.\n\nBefore the limited public release, Anthropic briefed senior U.S. government officials on Mythos' full capabilities, including staff at the Cybersecurity and Infrastructure Security Agency and the Center for AI Standards and Innovation, according to Bloomberg. The model's cybersecurity capabilities appear to have opened doors at the White House that the legal and political battle had kept firmly shut.\n\nAround April 15, 2026, Gregory Barbaccia, Federal Chief Information Officer at the White House Office of Management and Budget, sent an email to Cabinet department officials — including counterparts at the Department of Defense, Treasury, Commerce, Homeland Security, Justice, and State — indicating that OMB was establishing protections to potentially allow agencies to begin using a modified version of Mythos. In that email, Barbaccia stated: "We're working closely with model providers, other industry partners, and the intelligence community to ensure the appropriate guardrails and safeguards are in place before potentially releasing a modified version of the model to agencies."\n\nThe signal from Barbaccia's email was quickly complicated, however. An OMB spokesperson subsequently told Nextgov/FCW: "OMB is not giving access to anything to agencies," adding that "there are no policy changes and there is no OMB policy process happening on this issue." The apparent contradiction between Barbaccia's email and the spokesperson's denial underscored the fluid and politically sensitive nature of the administration's internal deliberations.\n\n<h2>Amodei at the White House — and What Came Next</h2>\n\nOn April 17, 2026, Anthropic CEO Dario Amodei traveled to the White House for a meeting with Chief of Staff Susie Wiles, Treasury Secretary Scott Bessent, and other senior administration officials to discuss the Mythos model. Both sides described the session positively. The White House characterized it as "productive and constructive" and called it an "introductory" meeting, according to CNN and Axios. An Anthropic spokesperson described it as "a productive discussion on how Anthropic and the U.S. government can work together on key shared priorities such as cybersecurity, America's lead in the AI race, and AI safety."\n\nPresident Trump, when asked about Amodei's White House visit on a runway in Phoenix, Arizona, said he had "no idea" about the meeting, according to CNBC. Trump had previously told CNBC's Squawk Box that Anthropic "tends to be on the left" but "we get along with them," and told reporters his administration had some "very good talks" with the company, saying the firm is "shaping up," according to The Hill.\n\nAccording to a source familiar with negotiations cited by Axios, both sides sought to wall off the Pentagon legal fight from how the rest of the government engages with Anthropic — effectively separating the unresolved DOD dispute from the broader question of civilian agency access to Mythos.\n\nAnthropic's maneuvering in Washington has not been purely legal. After the Pentagon designated the company a supply chain risk, Anthropic hired lobbying firm Ballard Partners — where White House Chief of Staff Susie Wiles previously worked — according to federal disclosures cited by CNBC.\n\n<h2>Inside the Administration: Competing Voices, One Emerging Direction</h2>\n\nThe Axios report published April 29, 2026 reveals the depth of frustration inside the administration with the ongoing standoff. One anonymous Trump adviser described the situation to Axios in blunt terms: "This is a big problem. Everyone's complaining. There's all this drama. So this got elevated to Susie to hear Dario out, determine what is bullsh-t and start to plot a way forward."\n\nAn anonymous administration official, quoted by RedState, offered a similarly candid assessment of where the two fronts of the conflict now stand: "There's progress with the White House. There's not progress with [the Department of] War."\n\nAn anonymous U.S. official, quoted by Axios, described Anthropic's strategy around Mythos with notable directness: "They're using this Mythos cyber weapon to find friendly ears in the government. They're succeeding."\n\nFor its part, Anthropic has sought to frame its position around both legal principle and constructive engagement. An Anthropic spokesperson said in a statement: "While this case was necessary to protect Anthropic, our customers, and our partners, our focus remains on working productively with the government to ensure all Americans benefit from safe, reliable AI."\n\n<h2>Why This Matters: The Broader Implications</h2>\n\nThe potential reversal carries significant implications beyond any single company's federal contracts. The supply chain risk designation applied to Anthropic was, according to Mayer Brown's legal analysis, the first such designation ever applied to an American company — a designation previously reserved for companies associated with foreign adversaries. Judge Lin's sharp language in her preliminary injunction underscored the legal novelty and constitutional stakes involved. In her ruling, Lin wrote: "Nothing in the governing statute supports the Orwellian notion that an American company may be branded a potential adversary and saboteur of the U.S. for expressing disagreement with the government."\n\nIf the White House proceeds with guidance or an executive action that effectively circumvents the supply chain risk designation — even without formally rescinding it — it would leave unresolved the underlying legal and precedential questions raised by the original designation. The D.C. Circuit's April 8 ruling, which left the Pentagon ban in place, remains operative for defense-related procurement. Any executive action would therefore likely apply primarily to civilian agencies, consistent with the split legal landscape already established by the courts.\n\nThe episode also highlights how rapidly AI capability can reshape political calculations. Mythos' documented ability to identify thousands of zero-day vulnerabilities across major operating systems and browsers — capabilities briefed directly to senior U.S. cybersecurity officials before the model's public release — appears to have been a decisive factor in elevating the conversation from legal adversaries to White House-level negotiations in a matter of weeks.\n\nThe situation also illustrates the competitive dynamics now playing out in the federal AI market. OpenAI's move to claim the $200 million Pentagon contract that Anthropic had been negotiating occurred while Anthropic was locked in legal proceedings — a reminder that the federal AI landscape is being actively reshaped in real time, with significant commercial and strategic consequences for all major model providers.\n\n<h2>What Comes Next</h2>\n\nAs of April 29, 2026, the White House is actively developing guidance and a draft executive action, according to Axios. The Pentagon's supply chain risk designation remains legally in effect following the D.C. Circuit's April 8 ruling. Anthropic's lawsuits challenging that designation are ongoing in federal court. The GSA has restored Anthropic's products to federal procurement schedules for civilian agencies. Negotiations between Anthropic and White House officials appear to be continuing, with both sides signaling a desire to move past the standoff — at least outside the Pentagon.\n\nWhether a formal executive action materializes, and whether it is broad enough to allow meaningful civilian agency adoption of Mythos, remains to be seen. The OMB spokesperson's public denial of any active policy process, issued even as Barbaccia's email circulated among Cabinet officials, suggests the administration has not yet landed on a unified or finalized approach. The legal proceedings add a further layer of complexity: any executive action that is perceived as undermining the ongoing litigation could draw scrutiny from the courts already engaged with this case.\n\nFor now, Anthropic holds a paradoxical position in the federal landscape — simultaneously barred from Pentagon contracts under a legally unprecedented designation, restored to civilian procurement schedules following a federal injunction, and actively negotiating with senior White House officials over its most powerful AI model to date.\n\nFor more tech news, visit our <a href=\"/news\">news section</a>.\n\n<h2>AI Tools, Productivity, and What It Means for You</h2>\n\nThe rapid evolution of AI policy — from executive bans to White House deal-making in under 60 days — is a reminder that the tools shaping your work and health are being decided at the highest levels of government right now. At Moccet, we track how AI developments translate into real-world productivity and wellness gains, helping you cut through the noise and make informed decisions about the technology you use every day. <a href=\"/#waitlist\">Join the Moccet waitlist to stay ahead of the curve.</a>", "excerpt": "The White House is developing a draft executive action that would allow federal agencies to bypass Anthropic's supply chain risk designation and adopt Claude Mythos, according to Axios. The move marks a potential 180-degree reversal just two months after President Trump directed all agencies to cease using Anthropic's technology. Negotiations between Anthropic CEO Dario Amodei and senior White House officials, including Chief of Staff Susie Wiles, appear to be driving the shift.", "keywords": ["Anthropic federal ban reversal", "Claude Mythos White House", "supply chain risk designation", "Trump Anthropic executive order", "federal AI policy 2026"], "slug": "white-house-executive-action-anthropic-ban-reversal-claude-mythos" } ```

← Back to Tech News

More Tech News

Iran War Halts Middle East Data Center Investment

OpenAI Misses Key Revenue, User Targets in High-Stakes Sprint Toward IPO - WSJ

Exclusive: OpenAI, Anthropic meet with House Homeland Security behind closed doors on cyber threats