Photo: Wikimedia Commons / UK Government
A new scandal is brewing involving Elon Musk, as a recent report shows that the Tesla CEO told Nvidia to ship thousands of AI chips to X and x.AI instead of Tesla. Musk has an explanation for this move, but this doesn't mean it will be easy to fight mounting criticism regarding his management style.
Elon Musk has become extremely unpopular outside his circle of die-hard fans, as he devotes less time to managing Tesla and more to shitposting on X. Many accused Musk of being distracted since he bought Twitter in 2022. Coincidentally, Tesla's share prices have trended lower since the acquisition was announced in April 2022. This irked investors, who started challenging Musk's position as Tesla CEO.
A new scandal is brewing after the recent layoffs controversy and the sudden strategy change at Tesla. A CNBC report claims that the Tesla CEO has told NVIDIA to prioritize shipments of AI processors to X and x.AI ahead of Tesla. CNBC saw emails circulating at Nvidia last December showing that Musk ordered Nvidia to redirect 12,000 of shipped H100 AI chips originally slated for Tesla to X instead. In exchange, an equal number of Nvidia GPUs slated for January and June delivery to X were to be redirected to Tesla.
The email messages predate the Q1 2024 earnings call when Musk announced that Tesla would invest $10 billion in computing power to train its AI this year. Musk said that he wants Tesla to become the leader in AI and robotics, a plan that requires an AI infrastructure with a huge amount of AI processors from Nvidia.
However, the emails obtained by CNBC suggest that Musk presented an exaggerated picture of Tesla's procurement to shareholders. The emails circulating among Nvidia staff also highlight discrepancies between Musk's comments during the first-quarter earnings call, chip bookings, and FY 2025 forecasts.
The new scandal made Tesla investors question whether Musk is fulfilling his duties to Tesla while also managing other companies. Musk and Tesla are embroiled in a series of lawsuits by investors accusing Musk of market manipulation and insider trading. The fact that Tesla faces sales declines and rough waters ahead doesn't help either.
The GPU controversy shows that Tesla is indeed competing with Musk's other companies, something that wasn't obvious until now. Musk indicated that he wanted more control over Tesla before growing Tesla into a leader in AI and robotics. In a January post, Musk said he "would prefer to build products outside of Tesla" if he does not have at least 25% voting control. The statement was characterized as "blackmail" by some of Tesla's longtime investors.
The Nvidia emails seem to show that Musk is already prioritizing GPU shipments toward his other companies. Besides the chips, Musk has also moved employees from Tesla, including machine-learning scientist Ethan Knight. At least four other former Tesla employees involved in Autopilot and big data projects at Tesla have also been moved to x.AI.
However, in a post on X, Musk denied that his decision to redirect the Nvidia chips to X and x.AI was detrimental to Tesla. On the contrary, the decision ensured that the chips were put to work in time, considering that Tesla's Giga Texas data center was incomplete. If the chips were delivered to Tesla, they would've just sat in a warehouse.
Musk revealed that the new data center's south extension at Giga Texas is almost complete. The building will house 50,000 Nvidia H100 GPUs at full capacity. The supercomputer is instrumental in training Tesla Full Self-Driving neural networks.
A new scandal is brewing after the recent layoffs controversy and the sudden strategy change at Tesla. A CNBC report claims that the Tesla CEO has told NVIDIA to prioritize shipments of AI processors to X and x.AI ahead of Tesla. CNBC saw emails circulating at Nvidia last December showing that Musk ordered Nvidia to redirect 12,000 of shipped H100 AI chips originally slated for Tesla to X instead. In exchange, an equal number of Nvidia GPUs slated for January and June delivery to X were to be redirected to Tesla.
The email messages predate the Q1 2024 earnings call when Musk announced that Tesla would invest $10 billion in computing power to train its AI this year. Musk said that he wants Tesla to become the leader in AI and robotics, a plan that requires an AI infrastructure with a huge amount of AI processors from Nvidia.
However, the emails obtained by CNBC suggest that Musk presented an exaggerated picture of Tesla's procurement to shareholders. The emails circulating among Nvidia staff also highlight discrepancies between Musk's comments during the first-quarter earnings call, chip bookings, and FY 2025 forecasts.
The new scandal made Tesla investors question whether Musk is fulfilling his duties to Tesla while also managing other companies. Musk and Tesla are embroiled in a series of lawsuits by investors accusing Musk of market manipulation and insider trading. The fact that Tesla faces sales declines and rough waters ahead doesn't help either.
The GPU controversy shows that Tesla is indeed competing with Musk's other companies, something that wasn't obvious until now. Musk indicated that he wanted more control over Tesla before growing Tesla into a leader in AI and robotics. In a January post, Musk said he "would prefer to build products outside of Tesla" if he does not have at least 25% voting control. The statement was characterized as "blackmail" by some of Tesla's longtime investors.
The Nvidia emails seem to show that Musk is already prioritizing GPU shipments toward his other companies. Besides the chips, Musk has also moved employees from Tesla, including machine-learning scientist Ethan Knight. At least four other former Tesla employees involved in Autopilot and big data projects at Tesla have also been moved to x.AI.
However, in a post on X, Musk denied that his decision to redirect the Nvidia chips to X and x.AI was detrimental to Tesla. On the contrary, the decision ensured that the chips were put to work in time, considering that Tesla's Giga Texas data center was incomplete. If the chips were delivered to Tesla, they would've just sat in a warehouse.
Musk revealed that the new data center's south extension at Giga Texas is almost complete. The building will house 50,000 Nvidia H100 GPUs at full capacity. The supercomputer is instrumental in training Tesla Full Self-Driving neural networks.
Tesla had no place to send the Nvidia chips to turn them on, so they would have just sat in a warehouse.
— Elon Musk (@elonmusk) June 4, 2024
The south extension of Giga Texas is almost complete. This will house 50k H100s for FSD training.
Google News