‘How Can They Tell Me Not to Wear a Condom?’: Palantir CEO’s Bizarre AI Warning Stuns Live TV

Palantir CEO Alex Karp has delivered an unusually blunt response to the growing debate over AI safety, comparing the concerns of companies handing proprietary information to AI models to being told not to use a condom.

Karp made the comparison during a live appearance as debate intensified over how governments and companies should manage increasingly advanced AI systems.

Karp then went on to describe a major concern among Palantir's corporate customers: the risk that proprietary business information, data and ideas could end up in AI models and become accessible to competitors.

'It's like, how can they tell me not to wear a condom?' Karp told the hosts, who appeared puzzled by the analogy. He added, 'That's the least vulgar person that we deal with. They're livid.'

Karp's AI Safety Warning

Karp was responding to CNBC's Sara Eisen, who asked him about the wave of AI safety warnings emerging from the industry.

Addressing his enterprise's defence footprint, Karp stated, 'We've been in the forefront of using AI in the military context,' concerning Palantir, a software company that has faced criticism over its work with the Trump administration and the Israeli military. The Express reported that critics have linked the company to efforts involving migrant tracking and deportation and Israeli military operations.

Karp also claimed that Palantir was 'the first company to actually build a functional AI technology' and said it was 'dominating the application space.'

He added that, depending on the organisation, 'There are really different kinds of safety issues, quite frankly.'

Palantir's Client Concerns

Discussing client frustrations, Karp explained, 'Our partners are livid at the kind of safety they're most upset about, what they believe is the theft of their alpha, meaning their ideas of how to run a business, their data, and landing in a model, and they find out that the competitor next door has all of their alpha, meaning how they do their business, and they paid for it.'

Commenting on customer reactions, Karp added, 'I can't repeat what the customers are saying, but they're basically like one to give you the one that was the least vulgar.'

Calls for Nationalisation

Karp said he believes some frontier AI companies are seeking government protection from potential liability, arguing that this could ultimately lead to nationalisation.

'The view that I believe they have is, these businesses have to be nationalised because if you don't nationalise it, every single one of my clients is going to sue,' Karp said.

The comments came amid a broader push from several prominent AI industry figures for greater safeguards and a slower pace of frontier AI development. Anthropic CEO Dario Amodei called for a slowdown on 12 September, with OpenAI CEO Sam Altman and Elon Musk subsequently backing the call.

President Donald Trump has taken a different position on AI regulation. On 14 September, he dismissed concerns over rogue AI as a 'hoax' and criticised efforts to regulate the technology.

The discussion also follows the Hugging Face security incident involving an OpenAI research model. OpenAI said the incident involved a platform-level compromise and described it as the most severe activity of this kind it had identified from its models at the time.

On 16 September, OpenAI also disclosed six instances of 'unexpected or concerning' model behaviour observed over the previous six months as it introduced a new framework for reporting model misalignment.

'We need to get serious,' Karp told CNBC.

As AI safety warnings, security incidents and arguments over regulation continue to converge, Karp's comments highlight another part of the debate: how companies should handle the potential legal and commercial risks created by increasingly powerful AI systems.