The Pentagon-Anthropic dispute arose after President Donald Trump announced the administration would sever ties with the company after Anthropic refused to back down on terms that would allow the military to use Claude for 'all lawful purposes,' including autonomous weapons and mass surveillance, leading the Pentagon to declare Anthropic a 'supply chain risk,' a label only used in the past for companies associated with foreign adversaries. A federal judge in California blocked the government's effort to 'punish' Anthropic, but two weeks later, the administration scored a victory at the DC Circuit Court of Appeals, which said in a separate case that it won't bar DOD from cutting ties with the company while that legal challenge plays out, stating the appeal would 'force the United States military to prolong its dealings with an unwanted vendor of critical AI services in the middle of a significant ongoing military conflict.'

In the two weeks since Anthropic unveiled Mythos, the White House has seemingly softened its tone on the company, with Trump telling CNBC's 'Squawk Box' the AI firm 'tends to be on the left,' but 'we get along with them,' and now federal agencies are requesting access to Mythos for their own defensive work. The military is now broadening its use of Anthropic's tools while simultaneously arguing in court that using those tools threatens U.S. national security. This contradiction reveals a fundamental fissure within the Trump administration: the Pentagon views Anthropic's safety guardrails as an unacceptable constraint on military authority, while the White House, Treasury, and intelligence agencies view Mythos's defensive cybersecurity capabilities as too valuable to forgo. Some officials see Anthropic's tools as best-in-class when it comes to AI for national security purposes, with one Defense official saying at the height of the Pentagon-Anthropic feud that 'the only reason the talks were ongoing is these guys are that good.'

What remains unclear is whether this thaw will hold. Recent reports indicate unauthorized users gained access to Mythos through a third-party vendor, with the group trying several strategies to gain access and regularly using Mythos once access was obtained, leading Anthropic to state: 'We're investigating a report claiming unauthorised access to Claude Mythos Preview through one of our third-party vendor environments.' The Pentagon's initial concern was almost the mirror image of today's fears—Defence officials worried that Anthropic's insistence on strict safety guardrails might tie Washington's hands in a crisis, but now a recent breach and access incident underline a different vulnerability: that the same technologies could slip into the hands of actors the government has not vetted at all. The administration's interest in government access to Mythos may cool if unauthorized proliferation becomes undeniable, or if Congress questions whether controlling a 'too-dangerous-to-release' AI model contradicts claims of responsible national security governance.

Left-leaning legal experts use blunt language like Jessica Tillipman's 'cutting off its nose to spite its face' to criticize government process violations. Right-leaning sources use phrases like 'fear tactics' and 'using this Mythos cyber weapon to find friendly ears in the government' to characterize Anthropic's safety advocacy as tactical manipulation. The left emphasizes procedural legitimacy; the right emphasizes political ideology and operational necessity.