Investors in Anthropic are actively working to reduce tensions between the AI startup and the Pentagon amid concerns that the dispute could significantly harm the company's business prospects, according to seven people familiar with the matter.
In recent days, CEO Dario Amodei has held discussions with several of Anthropic’s largest investors and partners, including Amazon.com CEO Andy Jassy, two of the sources said. Venture capital backers such as Lightspeed and Iconiq have also been in contact with Anthropic executives, two other sources added.
Those investor conversations have extended beyond the firm’s shareholder base. Some investors are reaching into their networks within the Trump administration in an effort to cool tensions, the sources said. The central focus of these outreach efforts, they said, is to avoid a designation that would require all Pentagon contractors to stop using Anthropic’s AI - a move investors warn could inflict severe damage on the company’s fast-growing enterprise business.
Anthropic and the Pentagon have continued some level of engagement while the dispute persists, according to one of the people. Reuters was unable to determine the substance of those continuing discussions.
The White House has publicly called on Anthropic to assist the government in phasing out its AI systems. The company and the Defense Department - which the Trump administration refers to as the Department of War - have been engaged in a months-long conflict over permissible military applications of Anthropic’s technology. The confrontation is widely interpreted as a test case for the degree of control AI developers can assert over the ways their systems are used across education, public services and other sectors.
At issue is whether companies like Anthropic will remove contractual red lines and instead adhere to an all-lawful-use clause favored by the Pentagon. Anthropic has repeatedly declined to permit its Claude AI to be used for autonomous weapons or for mass U.S. surveillance, refusing to abandon those prohibitions. The company was among the first in its peer group to handle classified information through a supply arrangement set up via cloud provider Amazon. OpenAI said last Friday that it had also reached a classified arrangement with the Pentagon and argued that Anthropic should not be singled out as a departmental risk.
Investor outreach and internal debate
During meetings with Anthropic executives, investors have voiced their continued support for the San Francisco-based lab while pressing for a path to resolution with the Department of War, the seven people said. Some investors expressed frustration that Amodei’s approach had antagonized Pentagon officials rather than building rapport.
"It’s an ego and diplomacy problem," one person familiar with the discussions told Reuters.
Investors face a difficult balance. Several said that Amodei cannot be perceived as capitulating to the administration without risking backlash from employees and customers who valued his stance. Amodei, who did not respond to a request for comment, has said Anthropic could not "in good conscience accede to their request." While speaking to investors late Tuesday, he said the company would "continue to work to figure out a solution with the DoW."
Funding and sales implications
Investors engaged in the outreach are focused on preventing the company from being labeled a "supply-chain risk" by the U.S. government - a classification that Defense Secretary Pete Hegseth has said would force all government contractors to cease using Anthropic’s technology. The imposition of such a bar could inflict what investors described as a potentially existential blow to Anthropic’s enterprise sales.
Anthropic has publicly countered Hegseth’s comments, arguing that he lacks the statutory authority to prohibit non-defense contractors from using its AI. The Pentagon did not respond to a request for comment on that claim. Anthropic also said on Friday it would challenge in court any supply-chain risk designation.
Some investors also worry that the controversy will deter private-sector customers who wish to avoid entanglement with the administration’s scrutiny. These concerns arrive at a pivotal moment for Anthropic: the company has said that enterprise sales account for roughly 80% of its revenue, and its future fundraising and any eventual initial public offering are likely to depend on continued growth in business revenue.
Anthropic has raised "tens of billions of dollars" on expectations for enterprise revenue growth, the people said. The company is in the process of permitting some employees to sell shares to investors and has previously indicated no final decision has been made about an IPO.
One of the people familiar with Anthropic’s financials said the company’s revenue run rate - the projected annual revenue based on current data - is about $19 billion, up from $14 billion a few weeks earlier.
Customer shifts and government action
The investor push to defuse the situation coincided with several U.S. agencies moving to end their use of Anthropic’s products. The State Department, for example, switched to OpenAI after President Trump issued an order on Friday directing agencies to remove Anthropic within the next six months.
Anthropic’s chatbot Claude and related tools such as Claude Code have seen marked demand: Claude was the most-downloaded free app in the Apple App Store on Monday, surpassing OpenAI’s ChatGPT, according to one of the people.
Still, investors say the possibility of a formal governmental restriction remains the gravest commercial threat. If a supply-chain risk designation were applied broadly, it could force government contractors to avoid Anthropic technology in any part of their operations, the people said. That outcome, they argue, would meaningfully constrain the company’s market access at a time when enterprise customers make up the bulk of revenue.
What the company says
Anthropic has disputed that the Defense Department possesses authority to block non-defense uses of its AI and has pledged to contest any formal exclusion in court. The Pentagon did not reply to requests for comment, and other investors, including Amazon, did not immediately respond to requests for comment.
For now, investors are pursuing diplomatic and political channels in hopes of limiting the dispute’s fallout while Anthropic and the Pentagon maintain limited engagement. The outcome of those efforts could shape the company’s ability to preserve enterprise relationships, support future fundraising, and sustain any path to an initial public offering.