Nvidia’s stock has been knocked around by shifting sentiment on artificial intelligence, but the core thesis behind the chipmaker’s meteoric rise has not changed. Wolfe Research is leaning into that view, arguing that the company’s AI engine still has years of growth ahead and that the recent pullback has left the valuation looking compelling rather than stretched. For investors trying to decide whether the AI trade is already over, the firm’s stance is a reminder that the rocket ship may simply be coasting before its next burn.
Why Wolfe still sees Nvidia as “very attractive”
I see the most striking part of Wolfe Research’s call as its refusal to treat Nvidia as a fad stock, even after a bruising stretch for AI names. The firm explicitly describes Nvidia’s valuation as “very attractive” in light of its AI driven growth prospects, a notable choice of words for a company that has already delivered enormous gains. In its latest work, Wolfe Research adds Nvidia (NVDA) to its recommended list and frames the recent underperformance as a function of seasonal trends rather than a structural break in demand, arguing that the long term earnings power from accelerated computing still justifies a premium multiple on Nvidia valuation.
That stance matters because it runs counter to the growing narrative that the AI trade has peaked. Instead of focusing on short term volatility, Wolfe Research is effectively telling clients that the company’s leadership in data center GPUs, networking, and software ecosystems like CUDA still has a long runway. By tying its bullish view to AI driven growth prospects rather than vague optimism, the firm is signaling that it expects hyperscalers, enterprise customers, and AI startups to keep building out infrastructure at a pace that supports Nvidia’s current scale and then some, even if quarterly results wobble along the way.
Rack scale forecasts show a long runway
Under the hood of that optimism is a very specific bet on Nvidia’s next generations of AI hardware. Wolfe forecasts about 55000 Blackwell racks and 20000 Rubin racks in 2026, a volume that would represent a massive deployment of high end AI systems into data centers worldwide. Looking further out, the same forecast calls for 55000 Rubin racks and 15000 Rubin Ultra racks in a later phase, implying that the product cycle is not just a one off spike but a multi year buildout of Blackwell and Rubin infrastructure.
In my view, those rack scale numbers are the clearest expression of why Wolfe believes Nvidia’s AI engine can keep running at high throttle. Selling tens of thousands of full racks, rather than just individual chips, locks Nvidia deeper into customer roadmaps and raises the switching costs for cloud providers that standardize on its architecture. It also supports the idea that AI workloads will continue to expand in size and complexity, from large language models to recommendation engines and digital twins, all of which benefit from the dense compute and networking that Blackwell, Rubin, and Rubin Ultra are designed to deliver.
Price targets and the debate over upside
Even as some investors question how much upside is left, top ranked analysts are still moving their targets higher. Earlier this year, one prominent voice lifted a formal price objective for Nvidia to $275, arguing that the company’s earnings power can keep compounding as AI infrastructure spending grows. That $275 figure is not a casual guess, it reflects detailed modeling of data center revenue, margins, and capital intensity, and it underscores how at least one influential analyst sees Nvidia’s current share price as leaving room for further appreciation in Nvidia NASDAQ NVDA.
Wolfe’s own work fits into that broader debate by effectively arguing that the market is underestimating the durability of AI demand. By calling the stock’s valuation very attractive and layering in aggressive rack forecasts, the firm is signaling that its internal price target embeds a multi year earnings ramp rather than a quick pop. For long term investors, the convergence of a $275 target from one camp and Wolfe’s bullish stance from another suggests that, while the easy money in Nvidia may be behind us, the consensus view on peak earnings could still be too conservative if AI infrastructure spending continues at a similar pace through 2026.
Addressing fears about AI sustainability
Behind the volatility in Nvidia’s share price is a deeper anxiety about whether AI demand can really stay this strong. According to one of Wolfe’s leading semiconductor analysts, Jan Caso, the stock has lagged at points for three key reasons, including the late launch of Blackwell and concerns about the sustainability of AI spending. Jan, According, Caso, Blackwell, and the other factors he cites have weighed on sentiment, with some investors treating any delay or hiccup in product rollout as evidence that the AI cycle is already rolling over, as highlighted in According Caso Blackwell.
I read Wolfe’s latest positioning as a direct rebuttal to those fears. By emphasizing that Nvidia’s valuation sits below its five year average on some metrics even after a powerful run, the firm is effectively arguing that the market has overreacted to timing issues and cyclical worries. The late launch of Blackwell may have shifted revenue between quarters, but it has not changed the fundamental need for more compute to train and run ever larger models. In that light, Wolfe’s stance is that investors should focus less on whether AI growth slows for a quarter or two and more on whether Nvidia continues to dominate the architectures that matter when the next generation of workloads arrives.
AI fatigue meets Wolfe’s “another scare” view
Zooming out to the broader market, there is no question that AI fatigue has set in for some investors. Bets on AI companies have dominated the US equity market for three years, and a growing number of traders are now wagering that the boom is over and that capital will rotate into more traditional sectors. That shift in positioning has contributed to sharp pullbacks in high profile AI names, including Nvidia, as investors question whether the extraordinary gains of the past few years can be repeated and whether the AI buildout has front loaded too much demand, a dynamic captured in the observation that Bets on AI have dominated trading.
Wolfe’s macro strategists, however, are pushing back on the idea that this is the end of the AI story. In a recent note, Chris Senyek at Wolfe argued that, Ultimately, “we view this as another AI scare with software and related areas bearing the brunt of it,” and he suggested that the sector could take another hit in February April before stabilizing. I interpret that as a call for patience rather than capitulation: if the current downdraft is just another scare, then the long term winners in AI infrastructure, including Nvidia, may emerge stronger once the speculative excess is wrung out. For investors who can tolerate volatility, Senyek’s framing of Ultimately Chris Senyek suggests that the current environment is less a bursting bubble and more a painful but necessary reset.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
