NVIDIA: 'The Party Is Going to End Soon' According to CNBC

"The skeptical panelist remains a "big believer" in being in the "second inning of AI," but trimmed NVIDIA because "there is no opportunity for them to change the narrative." The core friction: Hyperscaler CapEx is modeled by the Street to grow roughly 10%, while Jensen Huang is guiding to a 40% trajectory. That gap has to close somewhere."

"NVDA is up around 83% over one year and 1,352% over the past five years, with shares last around $215 as of Friday morning. As one panelist put it, "the run in the stock has already been there" since it traded at $16 in 2022."

"The counterpoint owns NVIDIA across multiple funds because "they've got proprietary technology" and "it's not expensive in my view." The fundamentals support that read. Q4 FY2026 revenue hit $68.13 billion, up 73% YoY, with Data Center Networking ripping 263% YoY on the NVLink ramp. Q1 FY2027 guidance sits at roughly $78.0 billion, explicitly excluding any China Data Center compute revenue. Forward P/E is 24x, with analyst consensus at $269.17."

"The panel's most important observation may be structural. "2 years ago, the issue with AI, with getting this going, was all about capacity and GPUs. Now it's about power. It's about infrastructure. It's about optics." Incremental dollars are rotating accordingly. Corning (NYSE:GLW) is up more than 330% over the past year on AI data center optical demand."

Investors focus on how long NVIDIA can sustain rapid gains. Bear concerns center on a mismatch between hyperscaler capital expenditure growth modeled around 10% and management guidance pointing to about 40%, implying the gap must narrow. NVIDIA shares have risen sharply over one and five years, intensifying worries that the stock run has already occurred. Bull arguments emphasize proprietary technology and a valuation viewed as reasonable, supported by strong revenue growth and Data Center Networking momentum. Guidance for the next quarter remains high, with forward valuation around the mid-20s multiple. A structural shift is noted: AI bottlenecks are moving from GPUs toward power, infrastructure, and optics, benefiting suppliers such as Corning.

Read at 24/7 Wall St.