The Pentagon is at odds with artificial intelligence company Anthropic over safeguards that would prevent the government from deploying its technology to target weapons autonomously and conduct U.S. domestic surveillance, according to a Reuters report.
This comes following extensive talks under a contract worth up to $200 million, with the U.S. Department of Defense and Anthropic coming to a standstill. The company’s position on how its AI tools can be used has intensified disagreements between it and the Trump administration.
Anthropic said its AI is “extensively used for national security missions by the U.S. government and we are in productive discussions with the Department of War about ways to continue that work.”
READ: Google, LinkedIn, and Anthropic leaders speak out on ICE shooting of Alex Pretti (
Anthropic representatives had raised concerns during discussions with government officials that its tools could be used to spy on Americans or assist weapons targeting without sufficient human oversight. The Pentagon opposed the tech company’s guidelines, saying they should be able to deploy commercial AI technology regardless of companies’ usage policies, so long as they comply with U.S. law.
Pentagon officials would likely need Anthropic’s cooperation moving forward according to reports. Anthropic’s models are trained to avoid taking steps that might lead to harm, and Anthropic staffers would be the ones to retool its AI for the Pentagon.
This isn’t the first time the Trump administration clashed with Anthropic over its safety guidelines. According to Semafor, last year Anthropic clashed with the White House over the company’s support for regulation of AI at the state level, opposing a law that would have preempted state AI regulation. The rift widened when Anthropic barred its models from being used for certain law enforcement activities.
READ: Will AI make coders obsolete? Anthropic CEO’s comments fuel H1-B visa row (
Amodei was among the tech leaders who recently condemned the fatal shootings of U.S. citizens protesting immigration enforcement actions in Minneapolis. The deaths have added to the concerns among some in Silicon Valley of their tools being used by the government for potential violence.
These discussions highlight complications in the relationship between Silicon Valley and the U.S. government, as the latest tech and AI continue to be increasingly used for military and intelligence purposes.
Anthropic is currently preparing for an initial public offering, in a race against rival OpenAI. The Financial Times reported that the IPO could take place as early as this year.
Amodei recently criticized both the U.S. administration and major chip companies at the World Economic Forum in Davos over the decision to allow sales of Nvidia’s H200 chips to approved Chinese customers. The remarks drew particular attention given Nvidia’s role as a major partner and investor in Anthropic.