Palantir CEO Alex Karp called for “reasonable guidelines” on artificial intelligence regulation, saying nationalization may be necessary as leaders ponder the risks of the technology.
“We have to find a way to set reasonable guidelines that are enforced, and that’s not possible,” he told CNBC’s “Squawk on the Streets” on Thursday. “The first line of defense is to take responsibility for your actions.”
Karp repeatedly cited the risks and responsibilities that come with building powerful technologies, calling for both the nationalization of AI laboratories and civil and criminal liability for technology developers “to be held accountable.”
“The view that I believe they have is that these companies have to be nationalized because if they don’t nationalize, all my clients will be sued,” he said.
The tech executive’s comments come as policymakers and technology leaders fiercely debate how to regulate advanced AI models after researchers warned of the technology’s dire risks to humanity. humanDario Amodei spoke out over the weekend, calling for a slowdown in remote areas.
Calls for AI regulation have brought together unlikely rival Amodei. OpenAIwith Sam Altman space x and tesla CEO Elon Musk. But President Donald Trump has pushed back in recent days, calling the risks of AI a “misinformation” in several social media posts.
Including technology CEOs Nvidiawith jensen fan meta platform CEO Mark Zuckerberg also joined him. “AI labs that can’t focus on collaboration will fall behind,” Zuckerberg told X this week.
OpenAI on Wednesday revealed six new cases of “unexpected or concerning” model behavior within the past six months. This comes on top of the Hugging Face breach, which sparked a debate about broader AI safeguards.
“We need to get serious,” Karp told CNBC on Thursday.
The AI alert has also sparked a legislative push in Washington state, prompting renewed calls for oversight.
Sen. Richard Blumenthal (D-Conn.), who introduced the AI Safety Act, told CNBC this week that the U.S. is “on the verge of completely losing control” and called for some form of “objective review” of AI models before they are made public.
Lawmakers are also pushing for the implementation of so-called kill switches, which would force AI labs to build capabilities to stop AI if the technology goes rogue. The bill has already been introduced in the House of Representatives. A similar move faces pushback in the Senate.
