Two years ago, companies like Meta and OpenAI were united against military use of their tools. Now all of that has changed.

At the start of 2024, Anthropic, Google, Meta, and OpenAI were united against military use of their AI tools. But over the next 12 months, something changed.

In January, OpenAI quietly rescinded its ban on using AI for “military and warfare” purposes, and soon after it was reported to be working on “a number of projects” with the Pentagon. In November, in the same week that Donald Trump was reelected US president, Meta announced that the United States and select allies would be able to employ Llama for defense uses. A few days later, Anthropic announced that it too would allow its models to be used by the military and that it was partnering with the defense firm Palantir. As the year ended, OpenAI announced its own partnership with the defense startup Anduril. Finally, in February 2025, Google revised its AI principles to allow for the development and use of weapons and technologies that might harm people. Over the course of a single year, worries about the existential risks of AGI had virtually disappeared, and the military use of AI had been normalized.

Part of the change has to do with the immense costs involved in building these models. Research on general-purpose technologies (the other GPTs) has often highlighted the importance of the defense sector as a way to overcome issues of adoption. “GPTs develop faster when there’s a large, demanding, and income-generating application sector,” economist David J. Teece wrote in 2018, “such as the US Defense Department’s purchases of early transistors and microprocessors.” The soft budget constraints and long-term nature of defense contracting, combined with the often blurry metrics of success, make the military a highly desirable customer for new technologies. Given the need of AI startups, in particular, to secure large and patient investments, the turn to military funding was perhaps inevitable. But this doesn’t explain the rapidity of the shift nor the fact that all the leading American AI research labs moved in the same direction.

...read more at archive.is