Analyst Dan Ives argues Wall Street still underestimates Nvidia ahead of Wednesday's fiscal second-quarter earnings, pointing to hyperscaler demand and an underpriced physical AI business. He says consensus models sit 15% to 20% below what the company will deliver over the next 12 to 18 months, while prediction markets already lean bullish into the print.
Dan Ives, now at Yorkville Ives, told CNBC that his skepticism runs against the crowd: he believes analyst models for the next 12 to 18 months sit meaningfully below what Nvidia will actually deliver. The call lands days before Nvidia reports fiscal second-quarter results after the market close on Wednesday, August 26, 2026.
Shares closed at $216.85, up 23.79% over the past year and 16.41% year to date. That run makes a bullish call harder to justify unless it identifies something the market has missed.
Hyperscaler demand backs the scale argument
Ives centers his case on two pillars: near-term hyperscaler and Asia exposure, and a longer-dated physical AI bet. The first has hard numbers behind it. Nvidia's last quarter produced $75 billion in data center revenue, up 92% year over year, with sovereign revenue up more than 80% year over year and infrastructure deployed across nearly 40 countries.
Hyperscalers are matching that pace. Colette Kress told analysts AWS plans to deploy more than 1 million Blackwell and Rubin GPUs starting this year, and Microsoft's Fairwater data center is already live ahead of schedule, running on hundreds of thousands of Blackwell GPUs.
Physical AI is the harder bet
Ives says the market gives Nvidia little credit for physical AI. "you're almost giving them minimal credit for physical AI", he told CNBC. Nvidia has disclosed that physical AI revenue exceeded $9 billion over the last 12 months, small next to data center revenue but growing. The bet assumes inference moves out of data centers and into machines operating in the world, a different demand curve than training clusters — though robotics and autonomous systems have taken longer to commercialize than advocates projected in prior cycles.
What Wednesday needs to confirm
Prediction markets on Polymarket assign a 0.966 probability that Nvidia beats consensus non-GAAP EPS and a 0.845 probability of a new all-time high by year-end. Analyst consensus runs 48 Buy, 10 Strong Buy, two Hold and one Sell ratings, with an average price target of $302.83.
For the Ives thesis to hold, Wednesday's report needs a Q3 guide that steps up clearly on Blackwell and Rubin, confirmation that Vera Rubin production shipments begin in Q3, and enough sovereign and demand color to support the $1 trillion in Blackwell and Rubin revenue visibility management has staked out through calendar 2027. Last quarter's Q2 guide of $91 billion, plus or minus 2%, excluded China data center compute revenue, so any restart of H200 shipments to China would land as pure upside rather than validation of Ives's thesis.
Source: 24/7 Wall St.
Trading involves risk.
Risk Disclaimer
