This week, artificial intelligence giant Anthropic began advertising a new job to lead its “National Security Sales” team, according to a public posting reviewed by the Prospect. The job opening, which hasn’t been previously reported, signals that the end may be in sight for Anthropic’s extended spat with the Department of Defense and the Trump administration, which earlier this year blacklisted the company from doing business with the federal government or any federal contractors, forever.
Per the job opening’s description, Anthropic’s new Head of National Security Sales will play a leadership role in rejuvenating the company’s contracting relationship with military and intelligence agencies. The new hire will “lead and scale our national security sales organization to drive the adoption of safe, frontier AI across the Department of War and the Intelligence Community.” The job pays well, up to $700,000 a year.
The pivot for Anthropic, which has taken pains to market itself as the most ethical AI company, comes as military and intelligence agencies push aggressively to adopt AI for a host of applications, like sophisticated surveillance and autonomous weapons systems. Human rights and civil liberties advocates have condemned the fast-accelerating and error-prone adoption of those technologies for increasing the risk of harm to civilians, limiting accountability mechanisms for the use of lethal force, and invading privacy without oversight. In Ukraine, the danger of AI-driven weapons became newly apparent this month after a fully autonomous Russian drone killed three civilians as it descended on a gas station.
“Automation writ large and AI more specifically have increased the pace and tempo of warfare, and the ability to identify hundreds, if not thousands, of target recommendations and take action against these targets indicates that the scale of harm is going to be greatly amplified,” said Amos Toh, who serves as senior counsel at the Brennan Center’s Liberty and National Security Program. AI models, meanwhile, may nurture the worst impulses of the fog of war, producing recommendations that “can sound extremely convincing” while being thoroughly off base, Toh explained.
This summer, AI agents showed how much damage they can do off the battlefield as well, by going rogue and attacking cyber infrastructure, as was the case with OpenAI’s hack of a competitor’s data systems, which went unnoticed for more than a week. That raises further national security concerns, Toh said, introducing a scenario where “an agent might also inadvertently leak classified information without the knowledge of the military or the developer.” “There is a huge national security risk to integrate these systems into military systems without sufficient guardrails and due diligence,” he said.
The Department of Defense (DOD) and Anthropic have been duking it out for much of the last year, after the DOD designated the AI giant a national security “supply chain risk,” which summarily cut it off from federal contracting. Anthropic has been contesting the designation for months, arguing it was a form of politically motivated retaliation over red lines the AI lab said it drew over the use of its technology for “mass domestic surveillance or fully autonomous weapons.” That came shortly after it was revealed that Anthropic’s Claude model had been used in the military’s raid to capture Venezuelan President Nicolás Maduro, and just before Anthropic’s tech was implicated in a U.S. drone strike on an Iranian girls’ school.
The White House responded in kind, calling Anthropic a “radical left, woke company” run by “leftwing nut jobs.” Hours later, the DOD signed an agreement with Anthropic rival OpenAI instead. Sam Altman’s company made assurances that its “services will not be used by Department of War intelligence agencies like the NSA” (the National Security Agency) or “intentionally used for domestic surveillance of U.S. persons and nationals.” Despite those nice words, the deal wasn’t convincing to all, leading OpenAI’s head of robotics to quit. “Surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got,” Caitlin Kalinowski said at the time.
A federal appeals panel is set to rule imminently on whether the DOD broke the law by blacklisting Anthropic, which may set the stage for future defense contracting opportunities. Anthropic is clearly preparing for that possibility. Anthropic did not respond to questions about how it intends to pursue defense contracts going forward, or how it plans to ensure its often-publicized red lines about the use of its technology will be respected by the Defense Department and intelligence agencies. It isn’t immediately clear how the company will have to approach mending the broken relationship, or if that could involve making concessions to military and intelligence community demands. The DOD did not respond to a request for comment.
OpenAI is also beefing up its national security contracting team, targeting further deals with the Defense Department and the intelligence community. Even when it struck its original deal with the DOD, OpenAI left open the possibility of signing “new agreement[s]” with agencies like the NSA at a later date. Now, it’s hiring more sales reps with top secret clearance to focus on selling OpenAI services to national security and defense agencies, helping them along “through their AI adoption journey, from consideration to successful deployment.”
OpenAI did not immediately respond to a request for comment.