Photo: Tesla
Elon Musk has countered a CNBC report alleging that he diverted Nvidia AI processors intended for Tesla to his social media company, X.
Musk responded on X, stating, “Tesla had no place to send the Nvidia chips to turn them on, so they would have just sat in a warehouse. The south extension of Giga Texas is almost complete. This will house [50,000] H100s for FSD training.”
The Tesla CEO further elaborated a few minutes ago, “Of the roughly $10B in AI-related expenditures I said Tesla would make this year, about half is internal, primarily the Tesla-designed AI inference computer and sensors present in all of our cars, plus Dojo. For building the AI training superclusters, Nvidia hardware is about 2/3 of the cost. My current best guess for Nvidia purchases by Tesla are $3B to $4B this year.”
CNBC’s report suggested Musk exaggerated Tesla’s procurement of Nvidia’s H100 AI chips during the company’s first-quarter earnings call in April. It claimed internal Nvidia emails indicated a significant portion of these chips was redirected to X, potentially delaying Tesla’s advancements in autonomous vehicles and robotics.
The report cited a December Nvidia memo indicating that Musk had redirected 12,000 H100 GPUs from Tesla to X. This reallocation, CNBC argued, could hinder Tesla’s development timelines due to the high demand and limited availability of these processors.
Musk’s statement on X aimed to clarify Tesla’s current infrastructure capabilities. He highlighted the nearing completion of the south extension of Giga Texas, which will soon be equipped to support 50,000 H100 GPUs for Full Self-Driving (FSD) training.
Despite CNBC’s claims, Musk has consistently emphasized Tesla’s commitment to leading in AI and robotics. On the April earnings call, he reaffirmed, “If somebody doesn’t believe Tesla’s going to solve autonomy, I think they should not be an investor in the company. We will, and we are.”