Worried about how much upside from here Nvidia can actually get, says Susquehanna's Chris Roland

CNBC TelevisionAbout 4 min readFeb 26, 2026Watch original
THE SUMMARYAI-generated

Key Concepts

  • NVIDIA’s Guidance & Valuation: Discussion of NVIDIA’s recent financial guidance, its current valuation, and potential for future growth.
  • Semiconductor Cycle: The current stage of the semiconductor industry cycle, characterized as being in early stages of a “massive cycle.”
  • AI Impact on Software vs. Hardware: The contrasting effects of Artificial Intelligence on software (valuation compression) and hardware (positive outlook).
  • Hyperscaler Silicon Development: Increasing investment by major cloud providers (hyperscalers) in developing their own custom silicon.
  • Optical Interconnects: Emerging area of growth within the semiconductor space.
  • ASIC (Application-Specific Integrated Circuit): Custom-designed chips for specific applications, particularly relevant in the context of AI.
  • TPU (Tensor Processing Unit): Google’s custom AI accelerator.

NVIDIA’s Strong Guidance and Market Adjustment

The discussion began with acknowledging the market’s absorption of considerable information, but expressing relative comfort with NVIDIA’s current valuation. The speaker believes the market needs to adjust its expectations to align with NVIDIA’s performance. Chris Roland, Senior Semiconductor Analyst at Susquehanna, was brought in to provide further insight. He characterized NVIDIA’s recent guidance as “a monster guide,” noting the initial concern regarding potential limitations due to supply constraints, but highlighting NVIDIA’s effective management of these challenges. Susquehanna maintains a $250 price target for NVIDIA stock.

Supply Chain Management & Upside Potential

Roland emphasized NVIDIA’s exceptional supply chain planning, stating they likely secured supply commitments “two years in order to garner supply.” While acknowledging potential limitations to further upside, he believes “there is still some room there,” but cautioned that “the dream of dream scenarios might be a little bit more difficult to reach.” The primary concern isn’t necessarily memory costs, but rather memory availability.

Divergence Between Analyst Views and Market Sentiment

The conversation shifted to the broader market context, specifically the divergence between analyst perspectives and market valuations over the past six weeks. Despite a challenging “bear market in technology,” semiconductors have performed relatively well. Roland noted that the analyst community has maintained a consistent outlook, and his own model hasn’t significantly changed despite market fluctuations. He attributes this consistency to a growing understanding that the industry is still in the “early stages of a massive, massive cycle.”

AI’s Impact: Hardware vs. Software

A key argument presented is the contrasting impact of AI on different layers of the technology stack. Roland stated, “Hardware, we’re now understanding how powerful it is that we still are in the early stages of a massive, massive cycle. And unfortunately, software is being eaten by AI.” This translates to valuation compression in the application layer (software) while hardware, particularly semiconductors, continues to demonstrate strong potential. He specifically highlighted the “optical interconnect space” and “AI power using semiconductors” as areas for future investment. Optical interconnects refer to technology using light to transmit data, offering higher bandwidth and lower latency compared to traditional electrical connections.

Competitive Landscape & Hyperscaler Investments

The discussion addressed the potential risk to NVIDIA’s dominance from emerging competition. Roland acknowledged that “hyperscalers are going to make investments in their own silicon.” He specifically mentioned Google’s TPU (Tensor Processing Unit) and Amazon’s increased capital expenditure (CAPEX) on infrastructure as evidence of this trend. He also predicted a collaboration between ARM and OpenAI to develop their own ASIC (Application-Specific Integrated Circuit). An ASIC is a microchip designed for a specific application, rather than general-purpose use.

However, Roland tempered these concerns by pointing out that even with these internal developments, hyperscalers are still experiencing consistent growth, as evidenced by their reported numbers. This suggests that while diversification of silicon sources will occur, NVIDIA will likely remain a significant player. He stated, “There will be a diversification outside of NVIDIA. But if you look at the numbers posted by the hyperscalers here, they’re still growing at a consistent company rate.”

Synthesis/Conclusion

The core takeaway is that NVIDIA remains a strong investment despite potential competitive pressures. The company’s effective supply chain management, coupled with the ongoing, early-stage growth of the semiconductor cycle driven by AI, positions it favorably. While hyperscalers are investing in their own silicon, their continued growth suggests NVIDIA will maintain a significant market share. The divergence between market sentiment and analyst views highlights the potential for NVIDIA’s valuation to adjust upwards as the market recognizes the strength of the underlying fundamentals. The emerging areas of optical interconnects and AI-powered semiconductors represent further opportunities for growth.

AI summaries can miss context or contain errors. Check important details against the original video.

Go a little deeper.

Have a question about this video? Load its transcript to open the video chat.