NextFin News - In a move that has sent ripples through the global financial markets this Monday, March 2, 2026, a leading Wall Street analyst has issued a provocative upward revision for Nvidia, suggesting the semiconductor giant’s growth trajectory is only in its nascent stages. According to The Street, Hans Mosesmann of Rosenblatt Securities has doubled down on his bullish stance, arguing that the market is fundamentally underestimating the "upside" potential of the Blackwell architecture and the subsequent "Rubin" platform. This bold projection comes as the tech sector navigates a complex geopolitical landscape under the administration of U.S. President Trump, where domestic energy production and AI sovereignty have become central pillars of national economic policy.
The core of this shocking take lies in the belief that Nvidia is no longer a cyclical hardware company but a foundational software-and-silicon ecosystem. Mosesmann posits that the transition from traditional data centers to "AI factories" represents a $1 trillion installed base replacement cycle that is currently less than 20% complete. This perspective challenges the prevailing narrative that AI capital expenditure (CapEx) among hyperscalers like Microsoft and Alphabet is nearing a plateau. Instead, the data suggests a shift toward sovereign AI, where nation-states are building localized compute clusters to ensure data security and technological independence, a trend accelerated by the current U.S. administration's emphasis on technological decoupling and domestic manufacturing incentives.
From a financial perspective, the numbers supporting this bullish thesis are staggering. In the most recent fiscal quarters, Nvidia has maintained gross margins above 75%, a feat previously thought unsustainable in the competitive chip industry. The "shocking" element of the current analysis is the projected earnings power; some analysts now see a path to $5.00 or even $6.00 in earnings per share (EPS) by 2027, which, at current multiples, would imply a market capitalization exceeding $5 trillion. This valuation is supported by the rapid adoption of the Blackwell B200 chips, which offer a 30x performance increase for LLM inference workloads compared to the previous H100 generation, effectively lowering the total cost of ownership (TCO) for enterprise customers despite the higher sticker price.
The impact of U.S. President Trump’s executive orders regarding energy deregulation has also played a silent but critical role in this upside. By streamlining the permitting process for nuclear and natural gas power plants, the administration has addressed the primary bottleneck for AI expansion: power availability. As data centers require unprecedented levels of electricity, the easing of energy constraints in the U.S. allows for a faster deployment of Nvidia’s power-hungry GB200 NVL72 racks. This synergy between federal energy policy and private sector innovation creates a "flywheel effect" that bolsters the bull case for domestic tech infrastructure.
However, the road to this projected upside is not without its hurdles. Investigative analysis into the supply chain reveals that while CoWoS (Chip-on-Wafer-on-Substrate) packaging capacity has improved, the reliance on high-bandwidth memory (HBM) remains a pinch point. Furthermore, the aggressive trade stances taken by U.S. President Trump regarding high-end silicon exports to adversarial regions have forced Nvidia to constantly redesign its "compliance-ready" chips. Yet, Mosesmann and other bulls argue that the demand vacuum created by these restrictions is being more than filled by the "Enterprise AI" wave, where Fortune 500 companies are moving beyond experimental pilots into full-scale production deployments of generative AI agents.
Looking forward, the trajectory for Nvidia appears to be decoupling from the broader semiconductor index (SOX). While legacy automotive and industrial chip sectors face headwinds, the specialized compute market is entering a "super-cycle" defined by the shift from general-purpose CPUs to accelerated GPUs. If the current pace of architectural innovation continues, Nvidia’s software stack, specifically CUDA, will act as a formidable moat, making it nearly impossible for competitors to displace them in the short term. The "shocking upside" isn't just about selling more chips; it’s about Nvidia becoming the operating system of the 21st-century global economy.
Explore more exclusive insights at nextfin.ai.
