Anthropic, an artificial intelligence firm preaching “safe AI,” has publicly criticized the Trump administration’s pro-innovation AI policies while quietly pursuing lucrative defense contracts.
Known for its chatbot Claude, Anthropic describes itself as a “safety-first company,” believing that “AI has the potential to pose unprecedented risks” — or “unprecedented benefits for humanity” if managed responsibly, according to its website.
Backed by prominent Democratic donor Dustin Moskovitz, a co-founder of Facebook, Asana and Open Philanthropy, Anthropic has deep ties to the Democratic Party. Moskovitz has contributed millions to Democratic campaigns, including those of Joe Biden and Kamala Harris. The company’s leadership includes several former Biden administration officials, making it a who’s who of Democratic insiders steering Anthropic’s policy agenda.
Former Biden National Security Council official Tarun Chhabra leads national security policy at Anthropic, former Biden economic aide Elizabeth Kelly serves as its head of beneficial deployments and Netflix co-founder and major Democratic donor Reed Hastings sits on its board. (RELATED: DAVID BLACKMON: Enviros Head To Memphis For Another Attack On American Industry)
Anthropic CEO Dario Amodei has been vocal in his criticism of President Donald Trump. In a now-deleted pre-election Facebook post, Amodei said that “Trump represents a serious and legitimate threat to the rule of law,” called him a clown and compared him to a “feudal warlord” who uses his power for personal gain rather than the national benefit.
Amodei admitted that he “considered saying nothing at all this election in order to maximize my ability to work both sides.”
On June 5, Amodei wrote an op-ed for The New York Times challenging Trump’s “Big Beautiful Bill” 10-year moratorium on states regulating AI. He instead suggested the government should require transparency in company practices and model capabilities.
That same day, Anthropic unveiled a “custom set of Claude Gov models built exclusively for U.S. national security customers.”
White House AI and crypto czar David Sacks said on the All-In Podcast that companies like Anthropic are pushing for “global AI governance.”
Sacks said they are using a “tried and true tactic of people who want to give more power to the government — to scare the population” into believing AI is too dangerous to be left up to the free market and must therefore be controlled by the government, which he warned would lead to a monopoly by just a few AI companies.
He said they do this by conflating the possibility of an “X-risk,” or the chance that an AI could be created that cannot be controlled or contained.
Sacks acknowledged that there is a nonzero chance this could happen, but warned the U.S. shouldn’t “hobble our own innovation, our own AI efforts in the name of stomping out every possibility of X-risk,” arguing that doing so could cause the U.S. to lose the AI race to China.
“I actually think that probably the greatest dystopian risk associated with AI is the risk that the government uses it to control all of us,” Sacks warned.
Despite its conflict with the current administration, Anthropic has been bidding for government contracts, including a partnership with Palantir to offer the use of Claude to the federal government for public sector applications, according to FedScoop.
Amodei told the Financial Times that he “doesn’t want a world in which AIs are used indiscriminately in military and intelligence settings” but is choosing to work with defense and intelligence in a manner that he said is responsible.
As the Trump administration moves forward with policies that oppose Anthropic’s stated goals, it remains unclear how the company will compete for contracts going forward. Anthropic has not responded to a request for comment.
