Half of xAI's co-founders have now left Elon Musk's AI startup

February 11, 2026:

Jimmy Ba, co-founder of Elon Musk's AI startup xAI, has left the company. Ba was one of xAI's twelve founding members and previously served as an assistant professor at the University of Toronto, where he studied under AI pioneer Geoffrey Hinton.

In his farewell post, Ba wrote that xAI's mission is to advance humanity on the so-called Kardashev technology scale. He predicts that recursive self-improvement loops—AI systems that improve themselves—could "likely go live" within the next twelve months. Despite these promising prospects at xAI, he says he wants "to recalibrate my gradient on the big picture."

With Ba's departure, six of xAI's twelve original co-founders have now left—half the core team. Previous departures include Igor Babuschkin (formerly at DeepMind and OpenAI), Yuhuai (Tony) Wu (formerly at Google and Stanford), Kyle Kosier (formerly at OpenAI), Greg Yang (formerly at Microsoft Research), and Christian Szegedy (formerly at Google).

February 10, 2026:

After Igor Babushkin, another xAI co-founder has announced his departure. Tony Wu, who was responsible for developing the company's foundational models and reasoning capabilities, reported directly to Musk. He joined xAI from Google when the company was founded in 2023.

Wu is leaving after xAI recently allowed the creation of deepfake nude photos for weeks, only backing down under pressure from authorities. His departure likely has nothing to do with the controversies, though—the AI developer thanks Elon Musk for his support in his farewell post, referencing "all those battles we have fought together." Babushkin left xAI back in August 2025 and started his own AI safety fund. His departure followed several controversies around the xAI chatbot Grok, which drew attention for far-right statements, among other things.

The timing is still notable. SpaceX recently announced a takeover of xAI that values SpaceX at one trillion dollars and xAI at 250 billion dollars. The likely reason: xAI generates almost no revenue on its own, even though developing and running its models costs billions. Without major progress, xAI appears to have no clear path forward (but a ton of compute).