Palantir CEO Alex Karp doubled down on his criticism of the frontier artificial intelligence labs on Monday, arguing that enterprises shouldn’t be forced to give up their intellectual property to work with model makers.
“We have people trying to drug addict us to a future they [frontier AI models] believe they control,” Karp said in an exclusive interview with CNBC. “Now, I’ve spent a lot of time with Dario [Amodei] and the Anthropic crew. They want to tell you we have to march into a future where we own nothing, where our businesses aren’t profitable, where none of us have jobs, and where our adversaries win.”
Instead, Karp said, enterprises should control their own models and work with companies like Palantir that offer an application layer that sits on top of a company’s stack, thereby allowing the business to keep the data in-house.
Anthropic and OpenAI have said over the last month that customer data is secure and isn’t used to train their models. OpenAI went as far as to say businesses must explicitly opt in if they want their data used for model improvement.
Karp isn’t buying it.
“Every enterprise we interact with, and that includes some of the biggest and most important government enterprises in the world, is saying, ‘Why would we tokenmaxx [and] pay people for something that’s not useful and then not control the means that allow us to advance our business while keeping the value of the business inside?” Karp said.
CNBC has reached out to Anthropic and OpenAI for comment.
Known for being outspoken and unapologetically brash, Karp was the first executive to take aim at OpenAI and Anthropic publicly in early July, blasting their token structure in viral comments on CNBC. His comments ignited a fiery debate, with many Silicon Valley and political leaders coming out in support.
“All-In” podcast co-host and tech investor Chamath Palihapitiya told “Squawk Box” in July that Karp “deserves a medal.”
“As Alex Karp put it, ‘What the technical customers want is control over their compute, their models, their data stack, and their alpha. They want to know they own the means of production, and it’s not being transferred to someone else.'” Microsoft CEO Satya Nadella wrote in a post on X on July 12. “The current regime does precisely the transfer Karp and companies fear.”
JPMorgan CEO Jamie Dimon joined CNBC’s “Squawk Box” on July 15 and said that while he didn’t hear Karp’s comments, “companies are going to be looking at how they spend their money” and looking for return on investment.
The debate has continued to rage on, and in late July, Palantir signed an open letter, alongside Nvidia, Microsoft and other companies, defending open-weight AI models as essential for national security.
Since then, investors and CEOs have shared that they are looking at ways to use open-weight and closed-source models, posing a potential threat to the large language models. Adding to that pressure is the rise of low-cost Chinese models that have been accused of distilling U.S. models.
Open-weight AI models are available for users to download, modify and run on their own infrastructure.
Karp said he doesn’t see China’s use of distillation to copy the U.S. models as unfair when the frontier models are basically doing the same thing.
“How do you think the models got their value? They distilled all the value of IP everywhere, including enterprise, everywhere,” Karp said. “Like we’re in a battle here. Those things have to work.”
Karp’s quest for the “global movement” of AI sovereignty as a core mission was echoed in his second-quarter letter to investors on Monday that accompanied Palantir’s earnings.
“Every organization in the world is awakening to the risks of handing the creators of the language models the keys to their institutions, of letting the models loose within their homes,” wrote Karp.
Palantir posted blowout earnings numbers that showed accelerating revenue growth in key areas. Overall revenue grew 93% over last year and the AI software company’s U.S. commercial business jumped nearly 150% in the second quarter.
