Nvidia Shares Climb on Bullish New Price Target

Price Targets

The stock advanced modestly in early trading, reflecting investor confidence in Nvidia’s extended revenue visibility through 2028. Rosenblatt analyst Kevin Cassidy’s upgrade follows management meetings with financial analysts, where the company reinforced its ambitious multi-year growth trajectory. The move places Rosenblatt among the most bullish voices on Wall Street, with the $325 target implying further upside from current levels as markets digest the chipmaker’s expanding artificial intelligence opportunity.

The upgrade extends a pattern of analyst optimism that has pushed price targets steadily higher throughout early 2026. Wedbush recently lifted its target to $300, Goldman Sachs moved to $240, and Bank of America now sits at $275, all citing similar themes around durable AI infrastructure spending and Nvidia’s technological leadership.

Core Revenue Projections Drive Upgrade

Rosenblatt’s updated financial model now incorporates Nvidia’s guidance for approximately $1 trillion in combined revenue from its Blackwell and Rubin GPU platforms spanning 2025 through 2027. This represents a significant expansion from earlier visibility into roughly $500 billion for the 2025-2026 period alone, effectively doubling the firm’s medium-term demand outlook.

The revised forecast includes an additional $25 billion in revenue expected during the second half of fiscal 2028, tied to three emerging product categories: Groq language processing units, Vera standalone systems, and the next-generation Rubin Ultra architecture. This extended roadmap underscores management’s conviction that the AI infrastructure buildout will sustain momentum well beyond initial market expectations.

Nvidia’s recent financial performance supports this bullish positioning. The company reported second-quarter fiscal 2026 revenue of $46.7 billion, up 56% year-over-year, followed by third-quarter revenue of $51.2 billion, representing 66% annual growth. Data center revenue, driven primarily by AI accelerator sales, continues to dominate the business mix and shows no signs of deceleration despite the massive scale already achieved.

At the company’s GTC conference in mid-March 2026, CEO Jensen Huang formally articulated the $1 trillion revenue aspiration, framing it around integrated AI hardware spanning GPUs, CPUs, networking equipment, and full rack-scale systems. Management emphasized that this outlook assumes continued strong demand not just for model training workloads, but increasingly for inference—the deployment phase where AI models actually serve end users at scale.

Inference Economics Become Central Thesis

Cassidy’s note highlights a critical shift in the AI investment narrative: the transition from training-focused infrastructure to inference-heavy deployments. Rosenblatt expressed conviction that as the AI market accelerates growth through inference-stage applications, Nvidia-based platforms will maintain industry leadership.

This view aligns with broader management messaging. Huang has repeatedly emphasized that the AI sector is entering an “inference inflection point,” where the economics shift from one-time cluster buildouts for foundation model training toward ongoing, high-volume compute for real-world applications. Nvidia positions its Blackwell and Rubin architectures as optimized for both workloads, with architectural improvements specifically targeting inference efficiency and total cost of ownership.

The inference thesis carries significant strategic implications. If AI revenue increasingly resembles recurring infrastructure spending rather than lumpy capital projects, it could justify higher valuation multiples and reduce concerns about demand cyclicality. However, inference is also where competition from custom silicon, application-specific integrated circuits, and hyperscaler in-house chips poses the greatest threat to Nvidia’s dominance.

Competitive Moat Faces Long-Term Tests

While Rosenblatt’s upgrade reflects confidence in Nvidia’s near-to-medium-term positioning, the $1 trillion revenue path embedded in the new price target assumes sustained technological leadership and pricing power through at least 2027. Markets are weighing this optimistic scenario against several structural challenges.

Major cloud providers including Microsoft, Amazon, and Google continue developing proprietary AI accelerators aimed at reducing dependency on external suppliers and improving economics for their own workloads. AMD has gained share in certain segments, and specialized inference accelerators from startups and established players alike are proliferating. The more AI inference becomes commoditized, the greater the risk to Nvidia’s premium pricing and gross margins, which remain extraordinarily high by semiconductor industry standards.

Geopolitical and regulatory factors also loom. Export controls have already constrained Nvidia’s China business, and further restrictions remain possible. The company’s market dominance in AI accelerators has attracted antitrust scrutiny in multiple jurisdictions, though no major enforcement actions have materialized to date.

Supply chain execution represents another variable. Nvidia must coordinate with TSMC on advanced packaging, scale production of increasingly complex multi-chiplet designs, and manage allocation across hyperscaler customers with enormous and growing appetites. Any significant production bottleneck or yield issue could delay revenue recognition and disrupt the carefully sequenced product roadmap underpinning analyst forecasts.

Valuation Debate Intensifies at $4.5 Trillion Market Cap

With Nvidia’s market capitalization hovering around $4.5 trillion, the company already ranks among the world’s most valuable enterprises. The stock trades at a forward price-to-earnings ratio that, while compressed from earlier peaks, still reflects expectations for exceptional growth relative to the broader market and semiconductor peer group.

Bulls argue the valuation remains justified given Nvidia’s near-monopoly position in AI training, expanding inference opportunity, software ecosystem advantages, and visibility into multi-year demand that few technology companies can match. The $1 trillion revenue guidance through 2027, if achieved, would represent one of the most dramatic growth trajectories in corporate history for a company of Nvidia’s scale.

Bears counter that much of this optimistic scenario is already priced in, leaving the stock vulnerable to any disappointment in execution, competitive pressure, or macroeconomic headwinds that constrain enterprise and cloud provider capital expenditure. The law of large numbers presents a mathematical challenge: sustaining 40% to 50% annual growth rates on a $200 billion revenue base requires capturing an ever-larger share of global technology spending.

Analyst consensus estimates have risen steadily as management has provided greater visibility into order books and multi-year platform commitments. Yet the clustering of price targets in the $240 to $325 range suggests the Street is converging on a similar set of assumptions about AI infrastructure spending, Nvidia’s market share trajectory, and margin sustainability. This consensus creates risk of crowded positioning, where any negative catalyst could trigger sharp sentiment reversal.

Future Hinges on AI Capex Durability

The central question facing Nvidia shareholders is whether the current AI infrastructure boom represents a durable, multi-year investment cycle or a more concentrated buildout that will eventually plateau. Management’s $1 trillion revenue guidance through 2027 firmly stakes out the former position, and Rosenblatt’s $325 price target reflects confidence in that view.

Hyperscaler capital expenditure plans support the bullish case. Microsoft, Amazon, Google, and Meta have collectively signaled data center spending in the range of $60 billion to $100 billion per company annually through 2027, with AI infrastructure representing the dominant category. Enterprise adoption of generative AI remains in early stages, and new use cases continue emerging across industries, potentially broadening demand beyond the initial wave of foundation model developers.

Yet history cautions that technology infrastructure cycles often unfold in waves of overbuilding followed by digestion periods. The current pace of AI chip procurement may reflect customers racing to secure supply and establish competitive positioning, rather than steady-state demand. If application-layer monetization disappoints or inference workloads prove more economical than training, the revenue trajectory could moderate faster than current Street models anticipate.

Nvidia’s product roadmap provides some insulation against these risks. The company has maintained an aggressive annual cadence of architecture updates, with Blackwell, Blackwell Ultra, Rubin, and Rubin Ultra all positioned to ship within a compressed timeframe. This rhythm keeps customers on an upgrade cycle and makes it difficult for competitors to close the performance gap. The integration of networking, CPUs, and software into complete platform solutions further deepens customer lock-in and expands addressable revenue per deployment.

Analyst Summary

Bull Case:

• $1 trillion revenue visibility through 2027 provides multi-year growth runway
• Inference acceleration expands addressable market beyond initial training buildouts
• Technological leadership and ecosystem lock-in sustain competitive moat
• Hyperscaler capex commitments remain robust with no signs of deceleration
• Product roadmap maintains annual refresh cycle preventing competitive catch-up

Bear Case:

• Valuation embeds optimistic assumptions vulnerable to execution or demand disappointments
• Custom silicon and specialized accelerators threaten inference market share
• Law of large numbers makes sustaining growth rates mathematically challenging
• Geopolitical and regulatory risks could constrain revenue or margins
• AI capex cycle may enter digestion phase sooner than consensus expects