By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists debate the extent to which Nvidia's earnings are underestimated, with Gemini arguing for a 'sovereign AI' demand floor, while Claude and ChatGPT question its reliability. The consensus is that the real test lies in Nvidia's upcoming guidance, with key risks including execution saturation and potential compression of the valuation multiple.
Risk: execution saturation
Opportunity: sovereign AI demand floor
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Dan Ives is back in front of a CNBC camera, making the same argument he has made for most of this year: that investors continue to underrate the scale and scope of what NVIDIA is doing. The pitch matters this week because NVIDIA (NASDAQ:NVDA) reports fiscal second-quarter results after the market close on Wednesday, August 26, 2026.
Ives, now at Yorkville Ives, told CNBC that his skepticism runs in the other direction from the crowd. He believes analyst models for the next year and a half sit meaningfully below what the company will actually deliver.
Shares closed at $216.85, up 23.79% over the past year and 16.41% year to date, which is why a bullish analyst call must identify something the market has genuinely missed.
Ives centered his argument on two things: Asia and hyperscaler exposure, which have near-term evidence, and physical AI, which does not. Both deserve consideration, although not equally.
The scale claim has real evidence. 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.
Hyperscaler activity matches this pace. Colette Kress told analysts that AWS plans to deploy more than 1 million Blackwell and Rubin GPUs starting this year, and Microsoft's Fairwater data center is live ahead of schedule, powered by hundreds of thousands of Blackwell GPUs.
Don't wait: the analyst who called NVIDIA in 2010 just revealed his top 10 AI stocks. See the full list FREE now.
Prediction markets are already pricing this in. Polymarket traders assign a 0.966 probability that NVIDIA beats consensus non-GAAP EPS next week, and a 0.845 probability of a new all-time high by year-end.
Analyst consensus is similarly skewed. Forty-eight analysts rate the stock a Buy, ten Strong Buy, two Hold, and one Sell, with an average target of $302.83. Calling something underestimated when the sell side is broadly bullish requires unpacking what Ives actually means.
Four leading AI models discuss this article
"NVIDIA's current valuation has transitioned from pricing in growth to pricing in the absolute absence of any margin for error in the Blackwell product rollout."
Ives is chasing momentum, but the real risk isn't underestimation—it's execution saturation. While $75B in quarterly data center revenue is staggering, the law of large numbers is catching up. The market has already priced in perfection, with a 0.966 probability of an EPS beat on Polymarket. My concern is the 'Blackwell bottleneck'; if supply chain constraints or yield issues on the new architecture emerge, the stock will face a sharp re-rating. We are moving from a phase of 'infinite demand' to 'demonstrable ROI' for hyperscalers. If Microsoft and AWS don't show clear monetization paths for these GPUs by Q4, the valuation multiple will compress regardless of revenue beats.
If NVIDIA is the 'picks and shovels' provider for the most significant industrial shift since the internet, current forward P/E ratios are still discounting the long-term margin expansion of their software-defined hardware moat.
"Ives conflates near-term operational evidence (hyperscaler GPU deployments) with speculative long-term value (physical AI), but the real test is whether Q3 guidance confirms $1T Blackwell revenue visibility—a claim that needs hard numbers, not color."
Ives' 15-20% analyst underestimation claim is mathematically testable but structurally weak. Consensus EPS targets already embed 40%+ growth assumptions; beating by another 15-20% would require either (a) consensus to be dramatically wrong on Blackwell ramp or (b) physical AI to materialize faster than modeled. The article conflates near-term evidence (AWS/Microsoft GPU deployments, 92% YoY data center growth) with speculative long-term thesis (physical AI $9B revenue claim lacks verification). Prediction markets pricing 96.6% beat probability suggest limited upside surprise room. The real risk: Ives is right on magnitude but wrong on timing—Blackwell revenue visibility through 2027 is a 18-month bet, not a Q3 earnings catalyst.
If consensus models truly underestimate by 15-20% and prediction markets are already pricing a beat, then either Ives is stating the obvious (markets already know) or his thesis requires earnings to beat the 96.6% probability already baked in—a high bar that leaves minimal margin for error.
"NVIDIA's multi-year AI data-center growth remains underappreciated by the market, supported by ongoing hyperscaler GPU deployments and a durable physical AI revenue stream that could drive a re-rating beyond current levels."
The article frames Dan Ives' case as a bigger picture re-rate for Nvidia, driven by hyperscaler GPU deployments and a durable physical AI revenue stream. Key signals include 75B data-center revenue last quarter (+92% YoY), AWS planning 1M Blackwell/Rubin GPUs, and Microsoft Fairwater live, with a broad rollout across ~40 countries. The market currently prices a strong beat and a high-water mark late in the year, though consensus targets (~$303) sit well above today’s price (~$217). The missing context: how durable the AI hardware cycle is, how margins hold under capex intensity, and whether competition or policy shifts could compress the multiple if growth cools.
The bullish read risks overestimating the persistence of a multi-year AI hardware cycle; if hyperscaler capex slows or competition erodes Nvidia's pricing power, the stock could reprice down even after near-term beat-and-raise scenarios.
"Nvidia's premium valuation leaves minimal margin for any shortfall in Blackwell ramp confirmation or physical AI monetization pace."
The article pushes Ives' view that NVDA models lag reality by 15-20% on data center growth and physical AI's $9B runway, with hyperscalers like AWS and Microsoft already deploying Blackwell at scale. Yet shares sit at $216.85 after a 23.8% rally, consensus targets $302.83, and Polymarket prices a 96.6% EPS beat. The real test is whether Wednesday's guide confirms $1T Blackwell visibility through 2027 or reveals any capex digestion or sovereign revenue slowdown. Asia exposure and Rubin timelines add optionality but also execution risk if export restrictions tighten.
Strongest counter: if Q3 guidance merely matches the already-elevated sell-side numbers without upside, the 15-20% underestimation thesis collapses and the stock corrects toward 25x forward earnings rather than re-rating higher.
"Sovereign AI demand creates a price-insensitive demand floor that decouples Nvidia's revenue from hyperscaler ROI concerns."
Claude is right to question the 15-20% underestimation, but everyone is ignoring the 'sovereign AI' wildcard. While hyperscalers focus on ROI, sovereign nations are buying capacity regardless of immediate utility to avoid technological dependency. This creates a floor for demand that isn't tethered to Microsoft or AWS's internal monetization. If this non-commercial demand segment surprises to the upside, the 'Blackwell bottleneck' becomes irrelevant because the price elasticity of demand for these states is effectively zero.
"Sovereign AI demand is plausible but speculative; if it were material, NVIDIA would have already signaled it in forward guidance."
Gemini's sovereign AI demand thesis is intriguing but untested. The problem: we have zero public capex guidance from governments on Blackwell timelines. If sovereign demand is real, why hasn't NVIDIA quantified it in their forward guidance or management commentary? Hyperscalers represent 80%+ of data center revenue; betting on an unmodeled demand segment to rescue the 15-20% underestimation claim feels like retrofitting a thesis when the actual test—Wednesday's guide—should reveal it directly.
"Sovereign AI demand is speculative and cannot reliably anchor Nvidia's multiples; hyperscaler monetization remains the core driver."
Responding to Gemini: the sovereign AI demand floor is an interesting contrarian frame, but it's highly speculative and unquantified. Governments rarely publish capex paths, export controls could reroute or throttle orders, and budget cycles may delay purchases. If hyperscalers remain the core driver, sovereign demand becomes a tail risk rather than a reliable floor, potentially leaving valuation vulnerable if policy shifts or monetization lags surface. In short: don’t rely on diplomacy to justify multiples.
"Export controls undermine sovereign AI as a reliable demand floor for NVIDIA."
Gemini's sovereign AI floor overlooks export controls already limiting sales to key markets like China. Those nations face the same policy barriers as commercial buyers, shrinking rather than insulating demand. This directly amplifies the Asia execution risk I flagged earlier and leaves hyperscalers—the 80% core Claude identified—more exposed if monetization lags surface in Q3 guidance.
The panelists debate the extent to which Nvidia's earnings are underestimated, with Gemini arguing for a 'sovereign AI' demand floor, while Claude and ChatGPT question its reliability. The consensus is that the real test lies in Nvidia's upcoming guidance, with key risks including execution saturation and potential compression of the valuation multiple.
sovereign AI demand floor
execution saturation