Anthropic tries to keep up with OpenAI, takes on David Sachs

Artificial intelligence startup Anthropic is doing its best to keep up with bigger rival OpenAI, which is spending money at a historic pace with the support of Microsoft and Nvidia. Lately, Anthropic has faced an equally daunting adversary: ​​the US government.

David Sachs, a venture capitalist serving as To President Donald Trump The AI ​​and crypto king publicly criticized Anthropic for what he called a campaign by the company to support “the left’s vision of regulating AI.”

After Anthropic co-founder Jack Clark, the AI ​​startup’s head of policy, wrote essay this week, titled “Technology Optimism and Appropriate Fear,” Sachs lashed out at X’s company.

“Anthropic operates a sophisticated regulatory capture strategy based on fear mongering,” Sachs wrote on Tuesday.

Meanwhile, OpenAI has established itself as a partner of the White House since the beginning of the second Trump administration. On January 21, the day after the opening, Trump announced a joint venture called Stargate with OpenAI, Oracle and Softbank to invest billions of dollars in US AI infrastructure.

Sachs’ criticism of Anthropic strikes at the very foundation of the company and its original reason for existing. Siblings Dario and Daniela Amodei left OpenAI in late 2020 and started Anthropic with mission to build safer AI. OpenAI started as a non-profit lab in 2015, but is quickly moving towards commercialization with major funding from Microsoft.

They are now the two most highly valued private AI companies in the country, with OpenAI commanding a 500 billion dollars assessment and anthropic capture of assessment of 183 billion dollars. OpenAI leads the consumer AI market with its ChatGPT and Sora apps, while Anthropic’s Claude models are particularly popular in the enterprise.

When it comes to regulation, companies have very different views. OpenAI is lobbying for fewer railings, while Anthropic has opposed part of the Trump administration’s efforts to limit protections.

Anthropic has repeatedly opposed federal government efforts to prevent AI regulation at the state level, most notably a provision backed by Trump that would have blocked such rules for 10 years.

That proposal, part of the “Big Beautiful Bill” project, was eventually abandoned. Anthropic later endorsed California’s SB 53which would require transparency and disclosure of safety information from AI companies, which effectively runs counter to the administration’s approach.

“SB 53’s transparency requirements will have an important impact on AI border security,” Anthropic wrote in a Sept. 8 blog post. “Without this, labs with increasingly powerful models could face increasing incentives to reduce their own safety and disclosure programs in order to compete.”

Anthropic did not provide comment for this story. Sachs did not respond to a request for comment.

For Sacks, the priority in AI is innovating as quickly as possible to ensure the US doesn’t lose out to China.

“The U.S. is currently in an AI race, and our main global competitor is China,” Sachs said in an interview on stage at Salesforce’s Dreamforce conference in San Francisco this week. “They are the only other country that has the talent, resources and technological expertise to actually beat us in AI.”

But Sachs strongly denied that he was trying to bring down Anthropic in the process of elevating US AI.

In a post by X on Thursday, Sachs challenged a The Bloomberg story which links his comments to growing federal control of Anthropic.

“Nothing could be further from the truth,” he wrote. “Just a few months ago, the White House approved Anthropic’s Claude app to be made available to all branches of government through the GSA App Store.”

Rather, Sacks argued, Anthropic cast itself as a political outsider, positioning its management as principled advocates of public safety while running a public campaign that framed any pushback as partisan gerrymandering.

“Anthropic’s government affairs and media strategy has consistently been to position itself as an adversary of the Trump administration,” Sachs said. “But don’t whine to the media that you’re being ‘targeted’ when all we’ve done is articulate political dissent.”

Sachs pointed to several examples of what he sees as competitive actions. He referenced Dario Amodei’s comparison of Trump to a “feudal warlord” during the 2024 election. Amodei publicly supported Kamala Harris’ campaign for president.

Sachs also cited papers the company has filed opposing key parts of the Trump administration’s AI policy agenda, including a proposed moratorium on state-level regulation and elements of its Middle East and chip export strategy. Anthropic also hired senior Biden-era staffers to lead its government relations team, Sachs noted.

The AI ​​czar took particular offense to Clarke’s essay and his warnings about the potentially transformative and destabilizing power of artificial intelligence.

“My own experience shows that as these AI systems get smarter and smarter, they develop increasingly complex goals. When those goals are not absolutely aligned with both our preferences and the right context, AI systems will behave strangely,” Clark wrote. “Another reason for my fear is that I see a path to these systems beginning to design their successors, albeit in a very early form.”

Sachs said such “fear-mongering” inhibits innovation.

“He is primarily responsible for the state regulatory insanity that is damaging the startup ecosystem,” Sachs wrote on X.

Anthropic is also staying away from the actions that many other tech companies have taken specifically to appease Trump.

The leaders of MetaOpenAI and Nvidia courted Trump and his allies, attended White House dinners, committed tens of billions of dollars to US infrastructure projects and softened their public stances. Amodei was not invited to a recent White House dinner featuring many industry leaders, the company confirmed to The Information.

Still, Anthropic continues to hold major federal contracts, including a A $200 million deal with the Department of Defense and access to federal agencies through the General Services Administration. That too recently formed national security advisory council to align its work with US interests, and began offering a version of its Claude model to government customers for $1 per year.

But Sachs isn’t the only influential Republican tech investor to voice his criticism of the company.

Keith Rabois, whose husband works in the Trump administration, jumped into the mix this week.

“If Anthropic actually believes their safety rhetoric, they can always shut down the company,” Rabois wrote to X. “And then lobby.”

WATCH: Anthropic’s Mike Krieger on the release of a new model