Nvidia Delays Next AI Chip "Blackwell" B200 Due to Design Flaw

Nvidia's next major AI chip, the "Blackwell" B200, may not ship in large quantities until next year due to a design flaw. According to The Information, Nvidia has informed Microsoft and at least one other cloud provider that the chips will take at least three months longer to produce than initially planned. The delay is the result of a design flaw discovered “unusually late in the production process,” according to two unnamed sources, including a Microsoft employee, cited by the outlet.

Impact on AI Cloud Landscape

The B200 chips are the successors to the highly sought-after H100 chips, which power a significant portion of the AI cloud infrastructure and have contributed to Nvidia’s status as one of the most valuable companies globally. Nvidia spokesperson John Rizzo shared with The Verge that production of the B200 chip is expected to ramp up in the second half of the year but did not comment further on the rumors.

Testing and Shipping Timeline

Nvidia is now conducting a new set of test runs with Taiwan Semiconductor Manufacturing Company. The company will not ship large numbers of Blackwell chips until the first quarter of next year. Microsoft, Google, and Meta have reportedly ordered “tens of billions of dollars” worth of these chips.

Industry Context and Future Plans

This report comes a few months after Nvidia announced that "Blackwell-based products will be available from partners" starting in 2024. The new chips are set to initiate a yearly release cycle of AI chips from Nvidia, as other tech firms, such as AMD, are also developing their own AI chip competitors.

Conclusion: Navigating Delays in AI Innovation

Nvidia’s reported delay of the Blackwell B200 chips underscores the challenges and complexities of AI chip development. As the company works to resolve the design flaw and meet high demand from major tech firms, the AI industry continues to anticipate the impact of these powerful chips on future AI advancements.