Elon Musk says Tesla 'not about to replace Nvidia' as EV maker develops chips for cars, robots

Tesla CEO Elon Musk announced that the company's custom AI5 chip, intended for vehicles and data centers, will be manufactured by Samsung in Texas and TSMC in Arizona, with a goal of "excess production." While Tesla continues to leverage Nvidia GPUs for AI model training, the AI5's tailored design is projected to deliver superior performance-per-dollar by optimizing for Tesla's specific requirements. This move underscores Tesla's vertical integration strategy in AI hardware, aligning with a broader industry shift towards custom silicon for efficiency, even as Musk's xAI remains a substantial Nvidia customer for its supercomputing infrastructure.

Analysis

Tesla (TSLA) CEO Elon Musk confirmed the company's custom AI5 chip will be manufactured by both Samsung (OTCPK:SSNLF) in Texas and TSMC (TSM) in Arizona, targeting "excess production" for vehicle integration and data center use. This dual-foundry strategy for the AI5, an evolution of Autopilot hardware, aims for radical simplicity by focusing solely on Tesla's specific requirements, potentially yielding a "best performance per dollar for AI maybe by a factor of 10."

Tesla, having transitioned from Nvidia's (NVDA) Drive chips in 2019, continues to leverage Nvidia GPUs for AI model training, possessing computing capacity equivalent to 81,000 H100 chips. This vertical integration into custom silicon, akin to Apple's (AAPL) strategy, allows Tesla to optimize the AI5 for its unique needs by eliminating legacy components, enhancing efficiency and potentially reducing costs.

Despite internal chip development, Musk clarified Tesla is "not about to replace Nvidia," with his AI startup, xAI, remaining a significant Nvidia customer for supercomputing infrastructure. This hybrid approach suggests Tesla optimizes custom hardware for specific in-house applications while relying on market-leading solutions for broader AI training. The market's mildly positive sentiment (0.2) and moderate impact (0.65) reflect a nuanced view of this strategic development.

Key Decisions for Investors

• Investors should monitor Tesla's AI5 chip production ramp-up and its impact on vehicle performance and data center efficiency, as successful vertical integration could enhance margins and technological leadership.
• Evaluate the long-term implications of Tesla's hybrid AI strategy, balancing in-house chip development with continued reliance on Nvidia for high-end training, for both TSLA and NVDA.
• Assess the competitive landscape for custom AI silicon, considering how Tesla's approach compares to other hyperscalers and its potential to disrupt traditional chip supply chains.