Palantir CEO Alex Karp accused AI labs of 'drugging' users months before Anthropic IPO
As Anthropic celebrates a record-setting IPO and growing influence in the artificial intelligence industry, comments made months earlier by Palantir CEO Alex Karp are drawing renewed attention. Karp had launched a sharp critique of frontier AI companies and accused firms like Anthropic and OpenAI of pushing businesses towards a future where they lose control of their data, models and intellectual property.
In a recent interview with CNBC, Karp argued that the enterprises should be wary of becoming dependent in AI model providers. “We have people trying to drug addict us to a future they [frontier AI models] believe they control,” Karp said. Referring to Anthropic CEO Dario Amodei and his team, Karp added that they were advocating for a future “where we own nothing, where our businesses aren’t profitable, where none of us have jobs, and where our adversaries win.”Palantir pushes AI sovereigntyThe remarks made by Karp were part of a broader campaign led by Palantir promising what its calls ‘AI sovereignty’, the idea that organisations should retain control over their own models, infrastructure and data instead of relying heavily on external AI providers.
As per Karp, the businesses should work with software layers like those offered by Palantir while keeping their core data and AI assets in-house. He also stressed that the companies increasingly want ownership of their computing resources, models and data rather than transferring value to third-party providers.The comments from Karp, came as Anthropic and OpenAI faced scrutiny over how enterprise data is handled.
Both the companies claimed that the customer information is secure and is not used to train their respective models unless the customers opt in explicitly. ChatGPT-maker OpenAI has also stated that businesses must choose to share data for model improvement programmes. Karp, however, said many enterprise customers remain concerned about retaining control over the technologies powering their operations.Debate spreads across Silicon ValleyThe criticism made by Palantir CEO quickly started discussion throughout the technology industry.
Venture capitalist and “All-In” podcast co-host Chamath Palihapitiya publicly backed Karp’s position, saying the executive “deserves a medal” for raising the issue. Along with this, Microsoft CEO Satya Nadella also weighed in on social media, echoing Karp’s argument that enterprises increasingly want control over their compute infrastructure, models and data stack. JPMorgan CEO Jamie Dimon later joined the debate, saying companies would ultimately focus on return on investment and carefully evaluate where and how they spend money on artificial intelligence technologies.Open-weight models gain supportersAlong with this, the discussion initiated by Karp also expanded into the debate over open-weight AI models.
In July this year, Palantir joined Nvidia, Microsoft and many other companies and signed an open letter arguing that open-weight models are important for innovation and national security. Open-weight models also enable organisations to download, run and modify AI systems on their own infrastructure, providing greater control than fully closed-source alternatives. The supports also believe that such models can reduce dependence on a handful of AI vendors, while critics contend that frontier-model developers continue to provide the most advanced capabilities.Alex Karp questions AI training practicesKarp also drew comparisons between concerns over Chinese AI companies and the way frontier models are trained.
As low-cost Chinese AI systems gained traction and faced acuaccusations of distilling US technology, Karp argued that leading AI companies had themselves derived value from vast amounts of intellectual property available across the internet and enterprise ecosystems. “How do you think the models got their value?” Karp said. “They distilled all the value of IP everywhere, including enterprise.”Anthropic's blockbuster IPOKarp's remarks have gained renewed attention following Anthropic's record-breaking initial public offering, which underscored the strong investor appetite for frontier AI companies.
Claude-maker Anthropic made its blockbuster debut on Wall Street with the company’s valuation reaching nearly $2 trillion. Shares of the company closed at $97.17, up by 17% from its IPO price of $83. Soon after the debut, Anthropic shared a statement from CEO Dario Amodei reaffirming the company’s core focus on safety, committing to pace AI training and deployment as needed to maintain human control while resisting business or political pressures, framing this as essential to avoid risks like misuse or loss of control.
“Today puts the company in a stronger financial position, and in doing so only strengthens our resolve to do the right thing,” Amodei said. “If we can show it’s possible to make the right choices on safety and still build an incredibly successful company, others in our industry will be inspired to do the same. We call this a race to the top,” he added.Fearless legend or grounded realist?
Rate yourself on 11 categories in the Great Indian Confidence Survey – find out in 3–6 min