Palantir CEO Karp: AI Shifts Power to Working-Class Male Voters

A CEO telling a business network that his technology will strip economic power from one bloc of voters and hand it to another is not a normal earnings-cycle message, which is why Alex Karp's latest CNBC appearance is worth reading past the headline. According to Futurism, the Palantir chief said AI 'disrupts humanities-trained' and 'largely Democratic' voters while lifting the economic power of 'vocationally trained, working-class, often male, voters.' He added that anyone who does not see this belongs in an 'insane asylum.'

The framing matters because Palantir is not a neutral seller of models. The company provides AI-powered surveillance and analytics to customers that include Immigration and Customs Enforcement and the Israeli Defense Force, per Futurism's reporting. When the CEO of that company casts AI as a tool that specifically redistributes economic power along gender, class and party lines, he is telling investors, employees and government buyers how he wants his product understood.

The house has been on this beat for a while. Cofounder Joe Lonsdale reportedly called in December for 'bringing back masculine leadership' and complained about 'feminine energy running our cities and our courts.' Palantir's UK CEO Louis Mosley, whose grandfather was Oswald Mosley, has framed AI as a way to gut what he calls the professional 'lanyard class,' arguing 'we don't need bureaucracy to control AI, we need AI to cut bureaucracy.' Karp's CNBC line reads as a continuation, not a one-off.

The honest caveat is that this is prediction dressed as analysis. Karp did not, in the reporting available, point to a specific study, dataset or model that shows AI displacing humanities-trained workers in the pattern he described. It is a political claim about who wins and who loses, delivered by an executive whose customer base benefits from that framing being taken as fact. Read it as positioning, not as a workforce forecast.

What is worth watching is whether other AI-vendor CEOs pick up the same lane. If they do, the buying signal for large enterprises stops being 'best model, cheapest inference' and starts being 'which vendor's politics do I want on my org chart,' which is a different market than the one most CIOs think they are shopping in.