(Reuters) - Nvidia (NVDA.O), opens new tab said the revenue opportunity for its artificial intelligence chips may reach at least $1 trillion through 2027, as the company outlined a strategy to compete more aggressively in the fast-growing market for running AI systems in real time.
At its annual GTC developer conference in San Jose, California, Nvidia unveiled a new central processor and an AI system built on technology from startup Groq - part of CEO Jensen Huang’s push to strengthen the company’s position in so-called inference computing.
Inference poses greater competition for Nvidia including from custom processors built by customers such as Meta, even as it dominates the market for chips used to train AI with its graphics processors.
"The inference inflection has arrived," Huang said.
He was speaking at a hockey arena with a capacity of more than 18,000 at the four-day conference that has become one of the biggest showcases of AI technology. "I just want to remind you, this is a tech conference," he told the audience.
Shares of Nvidia - the world's most valuable listed company with a market value of more than $4.3 trillion - briefly jumped on the $1 trillion forecast but pared those gains to close up 1.6%.
Huang did not offer more details on that prediction. But it marks a big step up from the around $500 billion revenue opportunity for 2026 Nvidia reiterated at its last earnings call.
"Huang mapping out a $1 trillion opportunity through 2027 underscores the durable demand for Nvidia’s AI infrastructure despite investor concerns," Emarketer analyst Jacob Bourne said.
"It signals Nvidia is sustaining its leadership in the AI chip market while the overall AI industry expands beyond early experimentation into large-scale deployment."
Huang said that inference, where AI systems answer questions or carry out tasks, will be split up into two steps. Nvidia's Vera Rubin chips will handle a first step called "prefill," where the user's request is transformed from human words into the language of "tokens" that AI computers use.
Groq's new chips will handle a second "decode" stage where the AI computer provides the answer the user is looking for.
IN-DEMAND AI CHIPS
Groq, a chip startup from which Nvidia licensed technology for $17 billion in December. Groq specializes in fast and cheap "inference" computing work, in which an AI model takes what it has already learned and uses it to answer a question or make a prediction in real time.
After spending hundreds of billions of dollars in recent years on chips for training their AI models, companies such as OpenAI, Anthropic and Facebook owner Meta Platforms (META.O), opens new tab are shifting toward serving hundreds of millions of users who are tapping those AI systems.
The keynote also touched upon a next-generation AI chip called Feynman, named after late American physicist Richard Feynman.
Huang also made the argument that part of Nvidia's competitive advantage was its CUDA chip programming software, which some analysts regard as its strongest shield.
"The installed base is what attracts developers who then create (the) new algorithms that achieve the breakthrough" technologies, Huang said. "We are in every cloud. We're in every computer company. We serve just about every single industry."
Analysts also expect Nvidia to elaborate on why it invested $2 billion each in Lumentum and Coherent, both of which make lasers for sending information between chips in the form of beams of light.
Despite that increased competition, some of which is coming from Nvidia's own customers designing their own chips, Nvidia remains central to the global AI ecosystem.
Nations such as Saudi Arabia are building custom AI systems for their own populations using its chips, and it is one of the only large U.S. companies that continues to release open-source AI software, a growing field of competition between the U.S. and China.
By Stephen Nellis and Max A. Cherney
March 16, 2026
