BackTECH
‘DeepMind is moving very fast, OpenAI is not on …’: When Musk wrote to his Neuralink associates
Emails in the Musk v. Altman trial reveal Elon Musk privately warned Neuralink colleagues that DeepMind was moving fast and OpenAI risked falling behind. The testimony shows Musk fixated on DeepMind's Demis Hassabis from OpenAI’s early days, including concerns over Hassabis’s leadership and strategy.
Why It Matters
The disclosures shed light on the strategic anxieties that shaped early OpenAI decisions and the competition landscape in AI research, potentially influencing views on governance and safety in AI development.
Timeline
3 Events
Musk email and trial revelations reported (May 6, 2026)
May 6, 2026
The article reports emails from Elon Musk to his Neuralink associates stating: 'DeepMind is moving very fast. I am concerned that OpenAI is not on a path to catch up. Setting it up as a non-profit might, in hindsight, have been the wrong move. Sense of urgency is not as high.' OpenAI president Greg Brockman testified that Musk raised Hassabis many times during OpenAI’s early years, calling him 'very consistent and fixated' on the man. Musk’s pre-OpenAI dinner questions included: 'Is Demis Hassabis evil?' After an earlier dinner with Hassabis, Musk emailed Brockman and Ilya Sutskever describing it as 'extremely alarming.' Hassabis reportedly declined Musk’s alternative to Google. The Verge describes Hassabis as a constant figure of fear among OpenAI leadership.
Alphabet AI infrastructure investment noted (2026)
2026
The article notes that Alphabet is putting roughly $175 billion into AI infrastructure in 2026 alone.
DeepMind funding crisis and Musk's competing pitch (early 2014)
2014
In early 2014, DeepMind was running out of money as venture backers pulled out, and Demis Hassabis needed a partner with serious compute and patience. Google’s Larry Page stepped in with a deal reportedly worth between $400 million and $650 million, according to The Verge. Elon Musk proposed an alternative to Google, warning Hassabis about the risks of handing AGI research to a corporation. Hassabis accepted the Google funding anyway, suggesting it could fund research at the scale required; Musk’s alternative could not guarantee the same. The article notes that the lab Musk feared went on to do exactly what he feared.