Nvidia expects to make US$1 trillion from AI chips through 2027
The company has pushed beyond its hallmark of graphics processing units which are used to train and run AI software
Add BT as a preferred source
DeeperDive is a beta AI feature. Refer to full articles for the facts.
[SAN FRANCISCO] Nvidia chief executive officer Jensen Huang, addressing crowds at the company’s biggest annual event, unveiled a variety of new products while predicting that its flagship artificial intelligence (AI) processors would help generate US$1 trillion in sales through 2027.
During a 2½ hour keynote address, Huang announced plans to push deeper into central processing units, Intel’s home turf, and introduced semiconductors made with technology acquired from startup Groq. The company even said that it was developing chips for data centres in outer space.
At the heart of Huang’s message: Demand for computing power continues to soar, and Nvidia is uniquely equipped to meet the challenge.
“I believe that computing demand has increased by one million times in the last two years,” he said. “It is the feeling that we all have. It is the feeling every startup has.”
Huang is contending with increasingly sceptical investors, who want more evidence that Nvidia’s booming sales growth will last. The trillion-dollar sales forecast, fuelled by orders from the company’s Blackwell and Rubin chips, offered evidence that demand remains solid.
Still, the outlook does not represent a major acceleration in sales growth. The company had previously forecast that data centre gear would bring US$500 billion in sales to the end of 2026. The latest forecast extends the outlook another year, doubling the cumulative amount.
Navigate Asia in
a new global order
Get the insights delivered to your inbox.
After initially rising as much as 4.8 per cent, the shares soon pared their gains on Monday. The stock was up 1.6 per cent to US$183.19 at the close in New York.
“The update should ease fears of a pullback in 2027 as Rubin enters the cycle, though it may also reset market expectations higher and raise the bar again,” Bloomberg Intelligence analyst Kunjan Sobhani said.
New products and partnerships were a highlight of GTC, an annual gathering that draws fervent crowds to San Jose, California. The new Groq chip is designed to boost the responsiveness of AI systems. The company also showed off a computer made up of general-purpose CPUs. That opportunity is “for sure” a multibillion-dollar business, Huang said.
SEE ALSO
The GTC rollout is Nvidia’s latest bid to promote AI computing and keep customers loyal to its technology. The company uses the event to announce partnerships with companies in a range of industries, aiming to show the increasing benefits of AI.
Nvidia discussed new or expanded pacts with companies such as International Business Machines, Hewlett Packard Enterprise and Adobe. It also strengthened ties with Uber Technologies, saying it was planning a fleet of Nvidia software-driven autonomous vehicles by 2028.
A flood of spending on AI chips has turned Nvidia into the world’s most valuable company. But it’s facing mounting competition from rivals such as Advanced Micro Devices (AMD), as well as its own customers attempting to produce in-house chips to handle AI.
Nvidia has accelerated its technology development in recent years, seeking to replace its entire product lineup on an annual basis while adding new components. The next design of its flagship AI processors, appearing in the second half of 2026, is called Vera Rubin, named for the pioneering astronomer whose observations provided evidence supporting the existence of dark matter.
Rubin will be followed by a generation named after Richard Feynman, the American physicist who died in 1988. It will have customised high-bandwidth memory, the company said, without giving further details.
While Nvidia continues to post sales growth that’s the envy of the chip industry, its stock rally has stalled in recent months. The shares were down 3.4 per cent this year heading into the GTC presentation, leaving the company’s market value at a still-unrivalled US$4.4 trillion.
Huang announced that the Groq 3 LPU will now be part of Nvidia’s product catalogue. An LPU, or language processing unit, is a specialised chip that’s good at accelerating large language model inference, the process of generating responses to AI prompts.
Such semiconductors have fast memory built onto the chip that helps them generate text almost instantly. Nvidia will offer it as a coprocessor, complementing work done by its accelerators. The latter components are better at dealing with more complex, multistage tasks.
In December, Nvidia announced what it called a licencing deal with Groq that gave it the rights to use certain technologies and designs. While the startup still exists, its founders and a big chunk of its engineers joined Nvidia in what was practically an acquisition.
Nvidia has accelerated the engineering work done by Groq to bring its product to market more quickly. The chip will be manufactured by Samsung Electronics, with Groq-based systems coming in the second half, Huang said.
The chipmaker said that its forthcoming CPU, branded Vera, is more capable than previous versions of that chip. As AI data centres become increasingly complex, the orchestration of work divided between various types of computers and software – a job performed by general-purpose CPUs – is becoming more important, Nvidia said.
Vera will combine the attributes of CPUs used in data centres, gaming PCs and laptops, the company said. It will be able to deal with many inputs simultaneously while quickly handling single complex ones. It also will require less electricity, according to Nvidia.
The company also plans to start selling computers made up entirely of the CPUs, a new approach for the chip giant. Such machines can be used in combination with other Nvidia-based computers or work independently.
Nvidia has pushed beyond its hallmark of graphics processing units, or GPUs, which are used to train and run AI software. It now offers complete computer systems featuring processors, networking and software.
The chipmaker also provides AI models and other software on an open-source basis, meaning that customers can tinker with the technology as they see fit. The company even offers versions tailored for specific uses, aiming to help industries that it sees as ripe for disruption by AI.
Nvidia got its start in data centre computing by offering versions of its computer graphics chips specially adapted to accelerate certain types of work. Over time that type of chip has overtaken the traditional dominance of the CPUs sold by Intel and AMD.
While that’s made Nvidia by far the largest provider of chips used in data centres, as software matures, many companies are exploring using CPUs to run the services that have been trained on the more specialised accelerator chips. CPUs are generally cheaper, can be used for other purposes and are usually less power-hungry.
Nvidia has so far only offered its CPUs tightly integrated with its other chips only. In a recent agreement with Meta Platforms, Nvidia indicated that it’s ready to sell its processors as a standalone product, and Huang has said repeatedly that a new version of the line will have greater capabilities.
A broader offering of processors by Nvidia may represent another challenge for Intel in what has long been its most lucrative business. It also brings in more competition for in-house efforts such as Amazon.com’s Graviton lineup.
SoftBank Group and its affiliate Arm Holdings have been expanding their capabilities in that area as well. Though Arm would benefit from Nvidia’s entry into the field – because it licenses technology to the bigger company – that income is potentially less than what would accrue from direct sales of its own designs and chips. BLOOMBERG
Decoding Asia newsletter: your guide to navigating Asia in a new global order. Sign up here to get Decoding Asia newsletter. Delivered to your inbox. Free.
Share with us your feedback on BT's products and services
