Nvidia’s next-generation AI chip roll-out slowed by engineering snags, say sources

Taipei – Nvidia hit engineering snags in the development of two new advanced chips, slowing the release of some products designed to extend its lead in the market for artificial intelligence (AI) computing.

The delays affected the company’s highly anticipated Blackwell line-up, which Nvidia announced in March, according to people familiar with the situation.

A version of the chip – known as an AI accelerator – is being reworked to better work with data centre infrastructure designed for an earlier chip, the Hopper H100. 

However, that is a relatively small segment of the market, the sources said. Separately, a product that combines a processor with a graphics chip will not be available in large numbers as quickly as hoped because of problems with supporting technology, they added.

The snags – first reported by the Information website – reflect the challenges of speeding up the pace of innovation at Nvidia. Chief executive Jensen Huang is introducing new chip designs and technology more rapidly, aiming to maintain an edge in AI computing.

The company dominates the market for AI accelerators – an advantage that has sent its sales and market valuation soaring over the past two years. 

Nvidia declined to comment on “rumours” about the recent engineering problems. The company has said it has begun sending out samples of Blackwell to customers broadly, and demand for its Hopper generation remains strong.

Blackwell “production is on track to ramp up in the second half”, Nvidia said in a statement.

Reports about the delays contributed to a 6.4 per cent stock decline for Nvidia on Aug 5, although a broader market rout also weighed on the shares. Meanwhile, rival Advanced Micro Devices gained 1.8 per cent – a sign investors hope it may be able to capitalise on the problems. 

Nvidia supplies its chips to companies such as Microsoft and Alphabet’s Google that are spending billions on building data centres, anticipating a surge in demand for AI services.

While the delays may affect the flow of the components, which are manufactured by Taiwan Semiconductor Manufacturing Co, analysts have mostly taken the concerns in stride. Given the acceleration of innovation, “bumps will continue to happen”, TD Cowen analyst Matt Ramsay said in a research note.

Weeks-long delays, if they materialise, probably will not have an impact on Nvidia’s rapid revenue gains or long-term growth, he said. But much will depend on how quickly Nvidia fixes the issues and gets chips to major clients. BLOOMBERG