Why Anthropic is Pushing Back Against US Government's AI Use

As AI is increasingly used for military purposes, a widening conflict has emerged between the US Department of Defense and AI company Anthropic over demands to remove safeguards from the company's AI systems, potentially reshaping how AI firms navigate government contracts and ethical boundaries.

Following the rift, US agencies have already begun phasing out Anthropic in favour of tools from OpenAI. 

Anthropic had been working with US government and military entities since June 2024, with its tools deployed across various operational levels including classified systems.

The company became the first frontier AI company to deploy models in the US government's classified networks, marking a significant milestone in AI-military collaboration.

This relationship fractured following pressure from US President Donald Trump and Secretary of Defense Pete Hegseth to enable "any lawful use" of the company's technology for military applications.

The company declined the request, arguing that existing legal frameworks have not evolved sufficiently to govern AI's capabilities.

Dario Amodei, Co-Founder and CEO of Anthropic, says in a public statement: "In a narrow set of cases, we believe AI can undermine, rather than defend, democratic values.

"Some uses are also simply outside the bounds of what today's technology can safely and reliably do.

"Two such use cases have never been included in our contracts with the Department of Defense and we believe they should not be included now."

The two applications Dario identifies are mass domestic surveillance and fully autonomous weapons systems

Government designation and legal challenge

After negotiations failed to achieve compliance, Pete designated Anthropic as a "supply chain risk".

This classification could mean organisations conducting business with the US military may no longer be permitted to engage with Anthropic.

The company responded, stating: "No amount of intimidation or punishment from the Department of Defense will change our position on mass domestic surveillance or fully autonomous weapons."

Anthropic announced plans to challenge the designation through legal proceedings.

"Designating Anthropic as a supply chain risk would be an unprecedented action – one historically reserved for US adversaries, never before publicly applied to an American company," the company says in a statement.

The designation represents a significant escalation in the dispute between the private sector AI firm and the federal government.

The company emphasised its commitment to supporting American military operations despite the conflict.

"We are deeply saddened by these developments," the statement says. 

"As the first frontier AI company to deploy models in the US government's classified networks, Anthropic has supported American warfighters since June 2024 and has every intention of continuing to do so."

Donald Trump posted on Truth Social directing all federal agencies to cease using Anthropic's services.

"We don't need it, we don't want it and will not do business with them again!" he said.

"There will be a six-month phase-out period for agencies like the Department of Defense who are using Anthropic's products, at various levels."

Trump's post escalated tensions further, stating: "Anthropic better get their act together and be helpful during this phase-out period or I will use the Full Power of the Presidency to make them comply, with major civil and criminal consequences to follow."

Privacy concerns and technical limitations

According to Dario, AI-driven mass surveillance "presents serious, novel risks to our fundamental liberties".

He questioned the legal adequacy of current surveillance frameworks, explaining that AI systems could be used to construct comprehensive profiles of individuals from disparate data points.

Dario pointed out in his statement: "Under current law, the government can purchase detailed records of Americans' movements, web browsing and associations from public sources without obtaining a warrant, a practice the Intelligence Community has acknowledged raises privacy concerns and that has generated bipartisan opposition in Congress."

Anthropic's position rests on the belief that "mass domestic surveillance of Americans constitutes a violation of fundamental rights".

The company maintains that enabling such capabilities would contradict its core mission of developing AI systems that respect democratic values and individual liberties.

While acknowledging that AI-enabled autonomous weapons could become necessary for national defence, Anthropic contends that current systems lack the reliability required for such critical applications.

According to Dario, autonomous weapons require safeguards that do not yet exist and these systems lack the critical judgement of experienced military personnel, potentially endangering both soldiers and civilians.

The company offered to collaborate with the Department of Defense on research and development to advance the technology for weapons applications, but this proposal was not accepted.

Dario's statement recognised it is the Department's right to select contractors aligned with their objectives, while expressing hope for reconsideration, which has now faded as OpenAI takes Anthropic's place. 

The feud with the US Government boosted Anthropic Claude's popularity and the app climbed to top of app store charts in US and UK, replacing the previous claimant – ChatGPT.