At CES 2026 in Las Vegas, Nvidia’s founder and CEO Jensen Huang announced that the OpenAI o1 model, released in September 2024, represents a watershed moment for artificial intelligence. Huang emphasized that AI is now evolving in a ‘thinking process’ rather than producing one-shot answers, driving exponential demand for higher-performance GPUs. He highlighted a 10x annual decline in cost for each generation of chips, underscoring how rapidly the industry is advancing and why computing speed has become the primary competitive battleground for AI developers.

Huang outlined how the precipitous price drop in previous-generation GPUs—Ampere, Hopper and Blackwell—will open a new market tier. As Rubin architecture chips launch later this year, the older models are expected to fall in price by roughly a factor of ten each year, mirroring patterns seen in consumer electronics. This dynamic will enable startups, mid-sized enterprises and academic labs to access powerful AI infrastructure at much lower entry costs, greatly expanding the developer base and solidifying Nvidia’s control over both high-end and mid-tier AI compute.

Nvidia disclosed that it has accumulated $500 billion in AI chip bookings through calendar 2026, with more than $150 billion already shipped. Fiscal Q3 2026 results (ended October 26, 2025) showed total revenue of $57 billion—a 62% year-over-year increase—while data-center revenue climbed 66% to $51.2 billion. CFO Colette Kress noted that full-year orders for the forthcoming Rubin family have already surpassed prior guidance, suggesting the booking figure will rise substantially by the close of fiscal 2026.

Analyst consensus forecasts call for Nvidia to generate $213 billion in revenue in fiscal 2026 and to reach $321 billion in fiscal 2027, representing roughly a 50% annual gain. Earnings-per-share estimates stand at $4.69 for the current year and $7.60 for the next. With open-source models like Meta’s Llama now responsible for approximately 25% of AI token volume, Nvidia’s GPUs continue to capture the lion’s share of AI workloads, positioning the company to extend its market leadership as enterprises worldwide step up data-center investments.