US President Donald Trump declared Friday that he was directing every federal agency to “immediately cease” deployment of Anthropic’s AI tools. This decision follows weeks of disagreement between Anthropic and senior officials concerning the military applications of artificial intelligence.
“The Leftwing nut jobs at Anthropic have made a DISASTROUS MISTAKE trying to STRONG-ARM the Department of War,” Trump stated in an update on Truth Social.
Trump indicated that there would be a “six month phase out period” for bureaus employing Anthropic, which might afford an opportunity for additional discussions between the administration and the AI startup.
The Pentagon and Anthropic offered no instant reply to inquiries for a statement.
The Department of Defense has endeavored to modify the conditions of an agreement reached with Anthropic and other firms last July to remove limitations on how AI can be applied, and rather authorize “all lawful use” of the technology. Anthropic opposed the modification, asserting that it could enable AI to completely command lethal autonomous weapons or to perform widespread oversight on US citizens.
The Pentagon presently abstains from employing AI in these ways, and has declared it harbors no intentions to do so. However, leading figures within the Trump regime have voiced opposition to the notion of a civilian tech company dictating defense deployment of such a vital form of tech.
Anthropic was the initial prominent AI research facility to collaborate with the US military, via a $200 million agreement inked with the Pentagon last year. It developed multiple bespoke systems known as Claude Gov that feature lesser constraints than its standard versions. Google, OpenAI, and xAI concluded comparable agreements concurrently, but Anthropic is the sole artificial intelligence firm presently engaged with classified systems.
Anthropic’s model is accessible through platforms supplied by Palantir and Amazon’s cloud platform for confidential military work. Claude Gov is presently primarily employed for ordinary duties, like composing accounts and condensing papers, but it is also utilized for intelligence analysis and defense strategizing, as per an insider acquainted with the circumstances who communicated with WIRED under the pact of remaining unnamed because they lacked official permission to disclose the affair publicly.
Over the past few years, Silicon Valley has shifted from mostly sidestepping defense work to progressively adopting it and ultimately evolving into complete military contractors. The conflict involving Anthropic and the Pentagon is now challenging the boundaries of that transformation. This week, several hundred workers from OpenAI and Google signed an open letter backing Anthropic and criticizing their respective firms’ choices to lift limitations on military use of AI.
In a memorandum dispatched to OpenAI staff today, CEO Sam Altman stated that the company concurred with Anthropic and also regarded mass surveillance and fully autonomous weapons as a “red line.” Altman further mentioned that the firm would endeavor to reach an accord with the Pentagon that would enable its ongoing collaboration with the military, The Wall Street Journal reported.
The open disagreement between the Pentagon and Anthropic commenced following Axios’s disclosure that US military leaders employed Claude to assist in formulating its mission to apprehend Venezuela’s president, Nicolás Maduro. Subsequent to the undertaking, a Palantir staff member communicated worries from an Anthropic employee to US military leaders regarding the deployment of its systems. Anthropic has refuted ever voicing apprehensions or interfering with the Pentagon’s utilization of its technology.
The contention involving Anthropic and the Department of Defense has intensified in recent days, with officials openly exchanging sharp remarks with the AI company across online platforms.
Defense Secretary Pete Hegseth convened with Anthropic’s CEO, Dario Amodei, earlier this week. He granted the company until Friday to pledge to altering the terms of its contract to permit “all lawful use” of its models. Hegseth applauded Anthropic’s offerings amidst the discussion and said that the Department of Defense aimed to persist in its collaboration with Anthropic, as per an individual cognizant of the exchange who lacked permission to publicly disclose it.
{content}
Source: {feed_title}