Nvidia faces key test at developer conference as AI market shifts to inference

Nvidia‘s developer conference has taken on heightened significance as the $3 trillion chip giant faces fresh challenges to its AI dominance. With pressure mounting on major customers to reduce the massive costs of AI infrastructure, CEO Jensen Huang must defend the company’s nearly 90% market share as competitors rush to develop more efficient alternatives. This market dynamic comes at a crucial inflection point, as AI computing shifts from training-focused workloads to inference applications that could reshape the industry’s competitive landscape.

The big picture: Nvidia faces a complex market transition as AI evolves from an era dominated by expensive training requirements to one focused on more diverse inference applications.

• The company’s revenue has more than quadrupled over the past three years to $130.5 billion, largely by selling computing power through high-end chips costing tens of thousands of dollars apiece.
• While Nvidia commands over 90% market share in AI training, the inference market—where AI models actually generate responses for users—is more contested territory with different technical requirements.

Competitive challenges: China‘s DeepSeek recently unsettled U.S. markets with a chatbot it claims requires significantly less computing power than rivals.

• This development spooked investors precisely because it threatens Nvidia’s core business model of selling extraordinarily powerful but expensive and energy-intensive processors.
• At least 60 startups along with established competitors like AMD are betting they can deliver inference computing at lower cost and power consumption.

Product roadmap: Nvidia is expected to reveal details of its next-generation “Vera Rubin” chip system at the conference.

• The new system, named after the American astronomer who pioneered dark matter concepts, is expected to enter mass production later this year.
• Meanwhile, its predecessor chip system named after mathematician David Blackwell has faced production delays that have eaten into Nvidia’s margins.

Strategic pivot: Nvidia argues that emerging “reasoning” AI capabilities will reinforce demand for its high-performance chips.

• These advanced models essentially “think aloud,” generating text and then reading it back to themselves to further consider problems—a process requiring substantial computing resources.
• “The market for inference is going to be many times bigger than the training market,” according to Jay Goldberg, CEO of D2D Advisory, potentially creating a massive opportunity despite lower percentage market share.

Beyond AI chips: Huang’s keynote will likely highlight Nvidia’s expansion into other computing markets.

• Following January comments where Huang suggested quantum computing was decades away from practicality—comments that crashed shares of quantum computing companies—Nvidia will devote an entire conference day to quantum technology.
• The company is also expected to provide updates on its personal computer processor development, first reported by Reuters and confirmed by Nvidia in January, which could further challenge Intel’s position.

Why this matters: The outcome of this market transition will determine whether Nvidia can maintain its extraordinary growth trajectory or if competitors can successfully carve out significant portions of the expanding AI computing landscape.

• As one startup executive put it: “They have a hammer, and they’re just making bigger hammers… every new chip they come out with has a lot of training baggage.”
• The fundamental question is whether Nvidia’s approach of building increasingly powerful general-purpose AI chips will prevail against more specialized, efficient alternatives tailored for specific inference tasks.