Nvidia CEO Jensen Huang has forecast a $1 trillion backlog in orders within the next year, as he outlined the company's vision for the next phase of the artificial intelligence boom. Speaking at a packed arena in San Jose, California, Huang described an 'inference inflection' that will drive demand for chips capable of running AI models after they have been trained.
Huang, 63, reiterated his belief that the AI buildup is still in its early stages, comparing it to the PC and internet revolutions. He predicted that Nvidia would face a $1 trillion order backlog by the end of the year, doubling his estimate from a year ago. The company's annual revenue has surged from $27 billion in 2022 to $216 billion last year, giving it a market value of around $4.5 trillion.
However, Nvidia's stock has cooled since briefly surpassing $5 trillion in market value last October, amid concerns that AI hype may be overblown. Even after a strong quarterly report in late February, shares remain 6% lower than before the results. Analysts expect revenue to exceed $330 billion in the coming year, but the company faces new competition from Google and Meta, which are developing their own processors.
Huang emphasised Nvidia's expansion into inference chips, which enable AI tools to produce responses more efficiently than training processors. To bolster this effort, Nvidia struck a multi-billion dollar licensing deal with startup Groq and hired its top engineers. Wedbush Securities analyst Dan Ives described the current period as a 'white-knuckle period' for tech, but predicted Nvidia's market value could surpass $6 trillion within a year.
