Trump Signals Potential Pentagon Deal with Anthropic as Tensions Ease

President Donald Trump has signaled a potential rapprochement with Anthropic, stating a deal for the Department of Defense to use its AI is "possible." This follows a period where the startup was blacklisted as a supply chain risk over disagreements regarding AI safety guardrails and autonomous weapons.

The back-and-forth continues in the relationship between Anthropic and the current U.S. administration. It seemed all was over for the AI startup regarding federal contracts after receiving a "supply chain risk" warning from the Pentagon. However, the launch of Mythos, an AI model allegedly so powerful it could pose a security risk if widely deployed, reopened the White House doors to the company. Now, President Donald Trump has indicated that Anthropic is "shaping up" and suggested that a future AI deal with the Department of Defense (DOD) is "possible."

The Trump administration is reconsidering Anthropic AI deal with DOD

The conflict between the company and the administration originally stemmed from a breakdown in negotiations over how the military could use the software. Anthropic reportedly sought guarantees that its AI would not power fully autonomous weapons or domestic mass surveillance. The Pentagon, seeking broader access for all legal purposes, resisted these constraints. This led to a breakdown in their $200 million contract.

The role of "Mythos"

As said before, the catalyst for this recent rapprochement seems to be Anthropic's new AI model called Mythos. Unveiled earlier this month, Mythos reportedly possesses unprecedented capabilities in identifying and exploiting cybersecurity vulnerabilities.

Because of its power, Anthropic has skipped a general release, instead launching "Project Glasswing." This initiative invites a select group of major tech companies, cybersecurity vendors, and financial institutions like JPMorgan Chase to evaluate the model privately. CEO Dario Amodei recently met with senior administration officials -- including White House Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent -- to discuss how Mythos could support U.S. cyber readiness and maintain the nation's lead in the AI race.

A potential middle ground

The President still occasionally uses sharp rhetoric regarding the company's internal culture. However, his recent comments emphasize a pragmatic interest in talent and capability. "They're very smart, and I think they can be of great use," Trump noted during his CNBC appearance.

The legal battle between the two sides continues in the background. Anthropic previously sued to reverse its blacklisting, resulting in a complex web of conflicting court rulings across various districts. However, the "constructive" nature of recent White House meetings suggests both parties may now prefer a negotiated settlement.