One AI company, Anthropic, is walking the walk. They’re planning to sue the Defense Department over its recent move to declare them a supply chain threat. This action comes amid U.S. military activity against Iran. It’s worth noting that this comes on the heels of a long and contentious battle over military establishment control of AI systems. This legal challenge will head to federal court, probably in Washington.
The Defense Department’s letter, which labeled Anthropic a supply chain risk, has drawn criticism from the company’s CEO, Dario Amodei. He claims the designation has no depth. For these reasons, it does a poor job of capturing the full picture about what’s going on with Anthropic’s operations. Amodei pointed out the memo was written within a few hours’ time. This swift action was immediately followed by a spate of statements declaring the opposite.
Anthropic continues to play a critical role in supporting U.S. operations, providing its AI models to the Defense Department at “nominal cost” for “as long as necessary to make that transition.” This commitment underscores the company’s priority to ensure that American soldiers and national security experts retain access to essential tools during ongoing major combat operations.
The new designation came after weeks of debate focused on how much control the military should have over AI systems. Amodei is convinced that the supply chain risk labeling only serves to shield the government from liability, not to penalize suppliers. He stated, “It exists to protect the government rather than to punish a supplier. In fact, the law requires the Secretary of War to use the least restrictive means necessary to accomplish the goal of protecting the supply chain.”
Noting this, Amodei recognized that the law makes it more difficult to challenge decisions like this one. The existing legislative framework makes it extremely difficult for companies to successfully challenge government procurement decisions. It provides the Pentagon sweeping leeway in issues involving national security.
In a somewhat ironic twist of fate, the Defense Department has just signed a contract with OpenAI to partner with them. This arrangement makes room for OpenAI to assume certain duties formerly shouldered by Anthropic. Amodei condemned OpenAI’s engagements with the Department as “safety theater.” He argued that their strategy doesn’t really serve to focus on the truly critical issues.
Amodei expressed apprehension about the potential ramifications of being given the supply chain risk designation. This would have a major effect on contractors doing business with the DoW. He clarified, “Even for Department of War contractors, the supply chain risk designation doesn’t (and can’t) limit uses of Claude or business relationships with Anthropic if those are unrelated to their specific Department of War contracts.” He noted that the memo has turned into an “obsolete judgment.” It’s too bad that it was written only six days ago, because it misses the realities we are now faced with.
In light of these developments, Dario Amodei stated, “It is not in our interest to escalate the situation.” He issued an apology for the confusion created by the sudden leak of the already-circulated internal memo. Further, he claimed that Anthropic was not negligent in how it did share it or in telling others to share it.
Anthropic remains dedicated to sustaining the U.S. national security mission. All while personally remaining on the frontlines of the government’s often treacherous regulatory maze and conflicting designations.