NextFin News - The era of triple-digit revenue surges and effortless market-cap milestones for Nvidia may be yielding to a more grounded reality, according to Sarah Kunst, managing director of Cleo Capital. Speaking in the wake of U.S. President Trump’s recent economic policy shifts and a high-stakes GTC conference, Kunst argued on March 21, 2026, that investors must now "normalize" their expectations for the semiconductor giant. The sentiment follows a period of unprecedented hypergrowth that saw Nvidia become the primary beneficiary of the global artificial intelligence infrastructure build-out, a cycle that Kunst suggests is entering a more mature, and perhaps more volatile, phase.
The shift in tone comes as Nvidia shares touched $188 earlier this week before retreating, a movement that reflects a broader market tension between AI optimism and macroeconomic caution. While CEO Jensen Huang used the recent GTC event to showcase the $1 trillion opportunity in inference chips—the hardware used to run AI models rather than just train them—the market response has been uncharacteristically muted. Kunst noted that while the fundamental strength of the company remains intact, the "shoes" Nvidia has created for itself are now so massive that even stellar performance can feel like a disappointment to a market conditioned for miracles.
Data from the most recent fiscal quarters supports this call for recalibration. While Meta recently announced a $115 billion AI spending bet that places Nvidia at the center of its multi-year cycle, the law of large numbers is beginning to apply friction to Nvidia’s growth curve. Maintaining a 200% year-over-year growth rate becomes mathematically grueling once a company reaches a multi-trillion-dollar valuation. Kunst pointed out that tech stocks often surge on momentum and can retract just as quickly when the narrative shifts from "limitless potential" to "execution at scale."
The competitive landscape is also thickening. Beyond the traditional rivalry with AMD, Nvidia now faces internal competition from its own largest customers. Hyperscalers like Amazon, Google, and Microsoft are increasingly deploying their own custom silicon to handle specific AI workloads, aiming to reduce their dependency on Huang’s expensive H-series and B-series chips. This diversification by big tech suggests that while the total addressable market for AI hardware is expanding, Nvidia’s near-monopoly on the most lucrative segments is facing its first sustained challenge since the AI boom began in late 2022.
Geopolitical factors under the current administration have added another layer of complexity. U.S. President Trump has maintained a rigorous stance on high-tech exports, ensuring that Nvidia’s most advanced Blackwell architecture remains restricted in certain international markets. These trade barriers, combined with a domestic focus on re-shoring semiconductor manufacturing, have forced Nvidia to navigate a labyrinth of regulatory compliance that adds cost and uncertainty to its long-term roadmap. Kunst suggested that these macro risks are often overlooked by retail investors who are still chasing the "moonshot" gains of 2023 and 2024.
The transition from training-heavy AI development to inference-heavy deployment represents the next major inflection point. If Nvidia can successfully dominate the inference market as it did training, the "normalization" Kunst speaks of might simply be a plateau before another ascent. However, the margin for error has vanished. In a market where a 4% rally can be erased in a single afternoon of profit-taking, the distinction between a great company and a great stock has never been more relevant for those holding Nvidia in their portfolios.
Explore more exclusive insights at nextfin.ai.
