Is safety “dead” in xAI?

Elon Musk is “actively” working on making xAI’s chatbot Grok “more disruptive,” according to a former employee who spoke to The Verge about recent departures from Musk’s AI company.

This week, after it was announced that xAI would be acquired by Musk’s SpaceX company (which previously acquired his social media company X), at least 11 engineers and two co-founders said they would leave the company. Some have said they are leaving to start something new, and Musk himself has suggested this is part of an effort to regulate cutting-edge AI more effectively.

But two sources who reportedly left the company (at least one of them before the current wave) told The Verge that employees had become increasingly disillusioned with the company’s disregard for safety, leading to global scrutiny after Grok was used to create more than a million sexual images, including fake images of real women and minors.

“Safety is a dead foundation at xAI,” one source said, while the other said Musk is “actively trying to make the model more disruptive because safety means oversight, to some extent, for him.”

They also reportedly complained about a lack of guidance, with one saying they felt xAI was “stuck in a catch-up phase” compared to competitors.