Economy April 13, 2026 03:42 PM

Anthropic Engages Trump Administration About Mythos After Pentagon Fallout

Co-founder Jack Clark says talks are underway despite a recent Pentagon contract termination and supply-chain designation

By Jordan Park
Anthropic Engages Trump Administration About Mythos After Pentagon Fallout

Anthropic is holding discussions with the Trump administration about its frontier AI model Mythos, co-founder Jack Clark said at a Washington event. The conversations follow the Pentagon's termination of business with the company after a contract dispute, and months after the agency labeled Anthropic a supply-chain risk and barred its use by the Pentagon and its contractors.

Key Points

  • Anthropic is in discussions with the Trump administration about its frontier AI model Mythos, as confirmed by co-founder Jack Clark at the Semafor World Economy event in Washington.
  • The Pentagon terminated business with Anthropic last month after a contract dispute and had in March labeled the company a supply-chain risk, barring its use by the agency and its contractors - developments that directly affect defense procurement and contractor relationships.
  • Anthropic announced Mythos on April 7, describing it as its most capable model for coding and agentic tasks; the specifics of which government agencies are involved in current talks and the content of those discussions remain unclear.

Anthropic has opened lines of communication with the Trump administration concerning its latest frontier artificial intelligence model, Mythos, the companys co-founder Jack Clark said on Monday. Clark made the remarks at the Semafor World Economy event in Washington, indicating the company is engaging government officials about its current and future models.

These talks come after a recent rupture with the U.S. Department of Defense. The Pentagon terminated business with Anthropic last month following a contract dispute between the two parties. In March the Pentagon had designated Anthropic a supply-chain risk, barring the agency and its contractors from using the company's services.

The underlying disagreement that prompted the Pentagon's action centered on the terms and guardrails for military uses of Anthropic's AI systems. Clark characterized the matter as a "narrow contracting dispute," while emphasizing the companys stated interest in national security and in ensuring government awareness of emerging models.

"We have a narrow contracting dispute, but I dont want that to get in the way of the fact that we care deeply about national security," Clark said. "Our position is the government has to know about this stuff ... So absolutely, were talking to them about Mythos, and well talk to them about the next models as well."

Anthropic unveiled Mythos on April 7, describing it as its most capable model to date for coding and for agentic tasks. Beyond Clarks statement that discussions are underway, the companies and officials involved have not disclosed which specific agencies are participating in the talks or the precise nature of those conversations.

The limited public details leave open several uncertainties about the trajectory of government engagement with Anthropic. The companys outreach to the administration follows an official prohibition by the Pentagon and a near-term contract termination, but Clarks remarks indicate Anthropic intends to brief government stakeholders on Mythos and subsequent models.


Contextual note: Public statements to date set out Anthropics intent to engage, the Pentagons earlier supply-chain designation and the companys April 7 announcement of Mythos, while the identities of participating agencies and the substance of the talks remain unspecified.

Risks

  • Uncertainty over which government agencies are engaged and the details of the discussions - this lack of clarity affects policymakers, defense contractors, and technology vendors that interface with federal procurement.
  • The Pentagons prior supply-chain risk designation and contract termination create regulatory and commercial headwinds for Anthropic in defense-related markets and among government contractors.
  • Disagreement over guardrails for military applications of Anthropics tools highlights ongoing tensions between AI developers and defense customers on acceptable use and contractual safeguards, with implications for adoption by defense and national security sectors.

More from Economy

Rising Fuel Costs Could Cancel Out Gains From Larger Tax Refunds, Morgan Stanley Warns Apr 15, 2026 Trump Urges Government Guardrails for AI in Banking While Citing Potential Benefits Apr 15, 2026 Pimco Acquires $400 Million of Blue Owl Bonds as Private-Credit Market Faces Outflows Apr 15, 2026 Trump Says China Will Stop Sending Arms to Iran as He Announces Permanent Opening of Strait of Hormuz Apr 15, 2026 Pakistani Delegation Set to Deliver U.S. Messages in Tehran, Iranian Spokesperson Says Apr 15, 2026