June 8, 2026, NVIDIA (NASDAQ: NVDA) traded in the $206–$210 range, with a market capitalization around $5.05 trillion and a static price-to-earnings (PE) ratio of roughly 31.8. However, its forward PE sits at just 20.9–21.6. This signals that the market expects earnings growth over the next 12 months to compress the current high valuation to a seemingly "normal" level. With about two and a half months until the next earnings report (August 25), we’re in a critical window for analyst forecast revisions and technology roadmap updates. Meanwhile, CEO Jensen Huang declined Senator Elizabeth Warren’s invitation to attend the Senate Banking Committee’s hearing on AI chip exports, placing NVIDIA once again at the intersection of geopolitical regulation and market expansion.
Breaking Down the Valuation Anchor: Is a 20.9x Forward PE Undervalued in the Age of AI Compute?
The core logic behind NVDA’s forward PE fluctuating between 20–22 isn’t about the absolute level of valuation, but rather how quickly earnings growth "dilutes" that valuation. For Q1 of NVIDIA’s fiscal year 2027 (ending April 26, 2026), revenue reached $81.62 billion, up 85% year-over-year. Non-GAAP EPS was $1.87, beating market expectations by 6%. Data center revenue contributed $75.2 billion, up 92%, accounting for about 92% of total revenue. Since Blackwell began mass production in the second half of 2025, NVIDIA has posted roughly 20% quarter-over-quarter growth for three consecutive quarters, demonstrating a stable demand flywheel effect. This translates to about $15 billion in incremental revenue each quarter, rapidly converting a lofty static valuation into "relative cheapness" from a forward-looking perspective.
However, the PEG ratio (price/earnings to growth) stands at about 0.52, which is a more telling metric. Generally, a PEG below 1 is considered a sign of undervaluation. Based on FactSet consensus, analysts forecast NVDA’s EPS to grow at an annualized rate of about 40% over the next five years. With a current forward PE of 21, the PEG is only about 0.53. The key assumption here is that the AI compute capital expenditure cycle won’t hit an inflection point before 2027–2028. If hyperscale cloud providers—Microsoft, Google, Amazon, Meta—slow their annual capex growth from the current 40–50% to below 20%, NVDA’s forward PE could quickly expand to 30–35 or higher, creating downward pressure on the valuation.
Morningstar’s target price of $280 implies about 35% upside, but this depends on two core variables: the production ramp of the Vera Rubin platform in the second half of 2026, and whether export controls further erode revenue from China. Thus, the current 21x forward PE is essentially the market trading at a "midpoint" price rather than a "bottom"—acknowledging structural demand for AI compute, while leaving room for geopolitical and competitive risks.
Dual Stress Test: Competitive Landscape and Export Controls
By 2026, the AI chip landscape has shifted from "NVIDIA’s sole dominance" to a three-pillar structure: NVIDIA leading, AMD and Broadcom each forging their own paths.
In the general-purpose GPU segment, AMD’s data center revenue hit $5.8 billion in Q1 2026, up 57% year-over-year. Yet, compared to NVIDIA’s $75.2 billion, the scale gap remains an order of magnitude. AMD plans to launch the Helios server rack with 72 MI450 GPUs in the second half of 2026, and has secured Oracle’s intent to purchase 50,000 chips. The effectiveness of this first large-scale deployment will be a decisive factor—any performance or stability issues could disrupt the fulfillment pace of potential orders.
Broadcom’s ASIC (Application-Specific Integrated Circuit) strategy shows cost advantages in inference scenarios. In fiscal 2025, Broadcom’s total revenue approached $64 billion, with custom AI chip orders from Google TPU, Meta, and Anthropic. If TPUs are successfully supplied externally in 2026, Broadcom could establish an inference-era moat similar to NVIDIA’s dominance in training. Inference workloads have different latency and efficiency requirements than training, creating room for ASIC differentiation. However, many complex inference tasks still rely on NVIDIA GPU software stack compatibility, so NVIDIA’s ecosystem moat remains difficult to bypass in the short term.
Vera CPU + Blackwell Architecture: The Technical Logic of a Supercycle
NVIDIA’s technology roadmap is evolving from a "GPU company" to a "comprehensive AI infrastructure platform." Currently, Blackwell architecture is midway through its production ramp, while the Vera Rubin platform—set for mass production in the second half of 2026—is seen as the key driver for the next product cycle.
Vera is NVIDIA’s first standalone data center CPU designed specifically for the era of AI agents. Based on an Arm architecture with 88 cores, it can decode 10 instructions per cycle in single-threaded performance, completing tasks about 1.8 times faster than traditional x86 CPUs. A critical shift: AI server CPU slots were previously dominated by Intel Xeon or AMD EPYC, generating $50–60 billion in annual revenue for those companies. With Vera CPU, NVIDIA is integrating the core controller of server hosts into its own product matrix, shifting the "host-accelerator" architecture from third-party CPU procurement to internal vertical integration.
The total addressable market (TAM) for standalone data center CPUs is about $200 billion. NVIDIA expects Vera CPU revenue visibility to approach $20 billion in 2026—a market segment previously excluded from most sell-side models. Given NVIDIA’s current annual revenue base of roughly $215.9 billion, even conservative estimates suggest this incremental $20 billion could add about 9% in extra revenue growth.
The Vera Rubin platform combines Vera CPU and Rubin GPU, delivering training performance 3.5 times that of Blackwell and 5 times higher AI software runtime performance. Inference throughput jumps 35-fold, driven by the sixth-generation NVLink switch chip’s revolutionary bandwidth upgrade. The significance: Even as unit compute prices fall, if performance gains outpace price declines, the "AI efficiency" per dollar continues to rise rapidly, incentivizing hyperscale customers to increase procurement rather than cut back.
Notably, Vera Rubin’s official shipment is expected in the second half of 2026. Whether its production ramp outpaces Blackwell’s early trajectory will be a key factor for fiscal 2027 revenue guidance. The ability to deliver next-gen technology on schedule is the biggest uncertainty in the long-term valuation logic.
Cross-Asset Mirroring: The Volatility Parity Phenomenon Between NVDA and BTC
In 2026, a notable statistical phenomenon emerged: The 30-day rolling annualized volatility for NVDA and Bitcoin is about 27.7% and 27.8%, respectively—a mere 0.1 percentage point difference. This volatility parity is much higher than the S&P 500 (about 14.9%) and Nasdaq 100 (about 18.4%), indicating that NVDA is being traded as a "proxy for crypto volatility" in the US equity market.
There are three transmission channels for this synchrony. First, macro liquidity transmission: When large tech stocks attract excess capital due to the AI infrastructure narrative, improved risk appetite flows into Bitcoin via institutional channels. Second, options market pricing linkage: NVDA’s put/call ratio volatility often leads Bitcoin’s implied volatility by 1–3 trading days, showing that the same cross-asset hedge funds are managing tail risk for both assets. Third, narrative overlap: AI chip demand is seen as the backbone of global digital transformation, while Bitcoin is viewed as the foundational protocol for digital value storage—both share the core narrative of "irreversible digitization" at different levels.
After NVIDIA’s May 20, 2026 earnings report, the company announced an $80 billion stock buyback and raised its quarterly dividend from $0.01 to $0.25. This capital return move directly boosted risk appetite, and Bitcoin’s price attempted to rebound to the $77,000–$78,000 range post-earnings. Cross-asset traders increasingly view NVDA as the "stock proxy" for the crypto market—when it rises, it’s seen as a macro signal friendly to crypto, and vice versa.
From a risk perspective, this synchrony means NVDA and Bitcoin share downside risk exposure. If macro liquidity reverses or the AI capex cycle peaks, both assets could face simultaneous capital outflows, rather than offsetting each other.
Gate Stock Trading: One-Stop Access to AI Chip and Crypto Market Synergies
Gate launched comprehensive US stock trading in 2026, covering core AI chip assets like NVDA, AMD, and AVGO, as well as spot Bitcoin ETFs and other crypto-linked assets. With NVDA and BTC volatility nearly equal, investors can allocate both asset classes on a single platform, managing cross-asset risk exposure seamlessly.
For trading access, users with Gate accounts can enable US stock trading and complete risk assessments, then search for NVDA and other ticker symbols directly on the Gate platform to place orders. Trades settle in USD, supporting fiat deposits and crypto collateral among other funding options. For investors tracking both NVDA’s valuation reset and Bitcoin’s price linkage, Gate offers an environment where all operations can be done without leaving the trading interface.
Conclusion
NVIDIA’s current $206–$210 share price and 21x forward PE reflect the market’s pricing of long-term structural demand for AI compute, while leaving room for geopolitical regulation and competitive risk. Whether the valuation is justified depends on Vera Rubin’s production ramp in the second half of 2026, changes in accessible Chinese market revenue under export controls, and the sustainability of hyperscale cloud capex cycles. The three competitors are pursuing differentiated technology paths—AMD focuses on general-purpose GPU catch-up, Broadcom on ASIC inference, and NVIDIA on full-stack vertical integration—each with strengths and weaknesses across training, inference, and customization, and no single approach has absolute dominance. Meanwhile, the rare volatility parity between NVDA and Bitcoin offers cross-asset investors a fresh analytical perspective.
The next key window for valuation recalibration is the August 25 earnings report. Gate’s stock trading platform provides users seeking a comprehensive investment view across AI chips and crypto markets with flexible, efficient tools and support.
