Fire in the Hole

Sep 12, 2:41 PM EDTbyNoor Al-Sibai

Elon Musk Appears to Now Be Firing the People Behind Grok

Something fishy is going on at xAI.

/ Artificial Intelligence/ Elon Musk/ Grok/ Twitter

Image by Jeff Kravitz / FilmMagic for HBO / Getty / Futurism

After vowing to create a “maximum truth-seeking” AI, Elon Musk appears to now be sacking the people who built it for him.

Slack screenshots leaked to Business Insider show that the accounts of at least nine high-level employees with the data annotation team at xAI, Musk’s OpenAI competitor, were deactivated over the weekend, in what seems like a major culling at the lab that built the social network’s infamous chatbot, Grok.

Those people, who weren’t named by BI, previously worked on the human data management team that oversaw the 1,500 or so workers and contractors that made up xAI’s “AI tutor” team, which is essentially tasked with training Grok how to be more intelligent for anywhere from $35 to $65 per hour, per the company’s career page . As BI reported earlier this year, those tutors have, to their chagrin, occasionally been asked to scan their faces for AI training purposes.

Sifting through LinkedIn profiles associated with the names of the employees whose profiles had been deactivated, the website identified roughly a dozen people who had worked on Grok’s human data managerial team. Many of them had posted on Slack as recently as September 5, the article notes.

Among those whose Slacks were deactivated was a supervisor who oversaw the rest of the managers, BI continued — and they worked on data annotation at Tesla’s Autopilot division prior to joining xAI.

Along with those Slack deactivations, multiple employees told the website that workers have been individually called in for meetings with supervisors. In those one-on-ones, the workers have been asked — much like when Musk was working in Donald Trump’s presidential administration earlier this year — what they’ve been working on and how they add value to xAI.

Those meetings, one worker old BI, have caused a “sense of panic” among the tutors.

xAI didn’t respond to BI‘s request for comment, a common practice.

Grok has been the subject of numerous scandals in recent months, including the “MechaHitler” debacle earlier this summer when the chatbot started targeting people deemed to have Jewish-sounding names, using racial slurs, and calling for a second Holocaust.

Were those managers fired for letting Grok go off the rails so badly? Or is Musk scaling down for some other reason? We’ll have to wait and see — but we’ll be watching.

More on Grok: Elon Musk’s Grok AI Spread Ludicrous Misinformation After Charlie Kirk’s Shooting, Saying Kirk Survived and Video Was Fake