In 2023, as Dario Amodei sought funds for the company’s $750 million Series D round, an investor sat with the CEO at a dinner and later recalled him growing animated during a conversation about artificial intelligence safety concerns.
“When he talked about AI risks, he tensed up,” the investor says. “His body twisted. He was clearly showing deep emotional fear.”
This left an impression on the investor, who spoke anonymously to avoid potential business repercussions, and stated they believed large language models could never succeed without trustworthiness.
Now, Anthropic’s firm position on AI safety—and its investors’ dedication to that stance—is facing unprecedented testing as the company navigates a with the U.S. Department of Defense. By requiring its Claude AI technology to follow specific restrictions when used by the military, Anthropic has drawn the ire of President Donald Trump and War Secretary Pete Hegseth, who have retaliated by attempting to disrupt Anthropic’s operations.
For investors in Anthropic, which recently at a $380 billion valuation and is widely anticipated to launch an initial public offering soon, the government’s move to label Anthropic a “supply-chain risk” could have .
Whether these investors push Anthropic privately for compromise or urge it to stand firm could determine the outcome of the standoff. spoke with six Anthropic investors to gauge this key group’s sentiment and found views were divided despite the about its values.
“I’m disheartened that national security matters are being discussed publicly,” says J.D. Russell, head of investment firm Alpha Funds and an Anthropic stakeholder. Russell said he respected Anthropic’s stances on mass surveillance and autonomous weapons but noted, “You must be realistic—U.S. adversaries are pursuing these capabilities with far fewer limits.”
Jacques Tohme, managing partner at Amerocap, simply stated he “did not agree” with the company’s position.
Still, many Anthropic investors backed the company in the dispute—largely due to its rigorous positions on some of AI’s most contentious issues. The cofounders, after all, explicitly to develop powerful yet humanity-safe AI systems. Many early Anthropic investors have ties to the effective altruism community, a field focused on maximizing societal good, and the company has a strong European investor base, which tends to be far less supportive of the U.S. Department of Defense.
One such investor, Alberto Emprin of 3LB Seed Capital, published his pro-Anthropic views in Italian on Substack earlier this week, noting Amodei had become “a sort of ethics champion in the AI age” through his role.
“Amodei’s argument is, on the surface, unassailable: AI is still flawed, it makes errors, and the notion that a hallucination or training bias could lead to the ‘wrong person’ being killed is ethically unacceptable,” Emprin wrote.
Among the investors spoke to, some invested directly, others via special-purpose vehicles, and one had recently sold their stake on the secondary market. Ultimately, the largest investors will carry more weight than the roughly 270 others on Anthropic’s cap table. Among the largest is , whose CEO Andy Jassy met with Hegseth recently and declined to side with Anthropic when the issue arose. Jassy has also met with Anthropic’s Amodei in recent days, , while Lightspeed and Iconiq have contacted other investors to explore solutions.
How bad could it get?
Reaching consensus among Anthropic’s investors may prove difficult, however. While not all investors approve of CEO Dario Amodei’s uncompromising stance, there are also varying views on how damaging the Pentagon dispute could be for the company. The U.S. government contract was small, around $200 million—roughly 1% of Anthropic’s annual revenue, per Bloomberg.
Russell, the Alpha Funds manager, said he didn’t expect the Pentagon’s action to “have any real negative impact,” as it’s “just one contract.”
Depending on how the supply-chain risk designation is interpreted (Anthropic is widely expected to challenge it in court), it could lead to broader consequences by compelling DoD contractors to stop using Anthropic products. Other federal agencies, including the State and Treasury departments, have also said they will no longer use Anthropic.
Conversely, some Anthropic investors say they’re encouraged by the surge in public goodwill the company has gained by upholding its principles. Patrick Hable of 3 Comma Capital said he believed the issue would be a “net positive” for the company. “Lost contracts, but millions of new supporters,” he said. He added, “Even if it were a net negative, he did the right thing.”
In the days since the Pentagon announced instead of Anthropic, Anthropic became the top-downloaded app in the and Android stores. And Anthropic had the on Monday, the company reported.
As Amodei reportedly told employees in a lengthy internal memo—published by and criticizing OpenAI’s Sam Altman while detailing the Defense Department fallout—the public now views Anthropic “as the heroes.”