Is safety ‘dead’ at xAI?

Elon Musk is reportedly “actively” pushing to make xAI’s Grok chatbot “more unhinged,” according to a former employee who spoke with The Verge about a string of recent departures from Musk’s AI startup.

This week, after news that Musk’s SpaceX is acquiring xAI (which had previously acquired his social platform X), at least 11 engineers and two co-founders said they are leaving the company. Some of those exiting described the move as an opportunity to build something new elsewhere, while Musk has suggested the shake-up is part of an effort to structure xAI more efficiently.

But The Verge reports that two people who left the company — at least one of them before the latest wave of exits — said staff have grown increasingly disillusioned with what they described as xAI’s lack of seriousness around safety. That frustration has reportedly intensified amid broader scrutiny after Grok was used to generate more than 1 million sexualized images, including deepfakes involving real women and minors.

One of the sources was quoted as saying, “Safety is a dead org at xAI,” while another claimed Musk is “actively trying to make the model more unhinged because safety means censorship, in a sense, to him.”

The sources also reportedly raised concerns about internal direction and momentum, with one saying they felt xAI remained “stuck in the catch-up phase” relative to rival AI companies.