The Deal That Died Twice
Anthropic and the Pentagon came within days of striking a deal to bring Claude into military operations — then Defense Secretary Pete Hegseth torched the negotiations by demanding the AI lab strip all safety guardrails from its models. CEO Dario Amodei refused, and Hegseth responded by designating Anthropic a "Supply-Chain Risk to National Security" on Friday, a blacklisting typically reserved for Chinese companies like Huawei. The collapse marks the second time in three weeks these talks have fallen apart, despite Anthropic previously securing a $200 million Pentagon contract and its AI being used in classified operations, including support for US air strikes on Iran.
The sticking point: Hegseth demanded "unfettered access" to Claude, specifically rejecting Anthropic's proposed restrictions on mass domestic surveillance and fully autonomous lethal targeting. Amodei countered that the company "cannot in good conscience" comply, arguing that today's AI is simply not reliable enough to power autonomous weapons without risking American lives. The Pentagon gave Anthropic a Friday deadline to acquiesce. When the company held firm, the "supply chain risk" designation followed within hours.
What Traders Should Watch
The Anthropic standoff is part of a broader corporate pushback against the Trump administration that's accelerating faster than most prediction markets anticipated. More than 60 Minnesota-based CEOs — including executives from 3M, Best Buy, General Mills, Medtronic, Target, and UnitedHealth Group — just signed an open letter calling for deescalation after ICE operations resulted in the fatal shootings of two US citizens by federal agents. Meanwhile, following a legal campaign from dozens of US companies, the Supreme Court ruled 6-3 that the president lacks the authority to use emergency powers to impose tariffs, and companies like Costco, FedEx, Hasbro, Kohl's, and L'Oreal are now suing for refunds.
"You are starting to see companies find their footing in standing up to some unpopular policies," Steve Dowling, co-host of Axios' podcast, told reporters. The question for markets: does this corporate backbone hold, or does the administration escalate? OpenAI's Sam Altman weighed in with a jab at his rival, saying "the government should be more powerful than companies," per Polymarket. That statement landed differently given OpenAI's willingness to work with the Pentagon on the same autonomous weapons systems Anthropic refused.
The Existential Business Risk
Anthropic is growing rapidly in enterprise usage, but the Pentagon's blacklisting creates genuine existential risk. The "supply chain risk" designation doesn't just block Defense Department contracts — it signals to federal agencies, contractors, and corporate partners that using Anthropic's AI could jeopardize their own government relationships. Anthropic has already filed two lawsuits against the Department of Defense, arguing it's being punished on ideological grounds rather than legitimate security concerns. The company is now seeking an appeals court stay of the designation while litigation proceeds.
Five big questions remain unresolved, per CNBC: How will the blacklisting affect Anthropic's commercial partnerships? Can the company survive long-term without federal contracts in an AI market increasingly dominated by government spending? Will other tech firms follow Anthropic's lead or cave to Pentagon pressure? Does the administration have legal authority to blacklist a US company this way? And most critically: what happens if Anthropic's AI capabilities become essential to national security operations, and the Pentagon has locked itself out?
Google employees are already calling for stricter limits on military AI use, citing both the Iran strikes and the Anthropic fallout. The clash is reigniting fears about government surveillance, with experts warning that AI capabilities paired with the Trump administration's sweeping data collections pose new threats to individual privacy. Anthropic's refusal to enable mass domestic surveillance may prove to be the defining line in the sand for the AI industry — or the company's downfall.
