Nvidia Up Only 7% This Year Significantly Underperforming Philadelphia Semiconductor Index, What Exactly Is Dragging Down This AI Computing King?

The legendary rally of the past few years has pushed Nvidia's market capitalization close to the $5 trillion mark, making Wall Street's scrutiny of its financial statements increasingly stringent.

In its fiscal first-quarter 2027 report released in May, Nvidia delivered stellar results with revenue doubling, but its stock price fell inexplicably the day after the earnings release, highlighting a severe desensitization of bullish sentiment in the market.

When expectations are pushed to the absolute limit, any positive news that merely meets expectations is interpreted as 'sell the news,' and Nvidia's elevated valuation premium suffered a ruthless correction in the first half of the year.

Although Nvidia's earnings continued to grow, the magnitude of its earnings beat hit a multi-quarter low. Data center revenue fell slightly short of the market's exceptionally high expectations, triggering investor concerns that AI chip demand is entering a 'high-level slowdown phase.' The R&D and launch of next-generation products drove up costs, and gross margin showed signs of narrowing, which was viewed as a slowdown in marginal profitability improvements. While its guidance came in above consensus estimates, it was still interpreted as 'lacking sufficient surprise' against the backdrop of sky-high expectations.

Under these circumstances, the earnings release instead became an opportunity for investors to take profits, while other companies in the sector, starting with lower expectations, found it easier to deliver positive surprises and naturally outperformed Nvidia.

Nvidia's underperformance against the Philadelphia Semiconductor Index is not due to a weakening semiconductor industry; on the contrary, it is because market capital is diffusing from "pure-play AI chips" to broader semiconductor enablers.

Memory chips were the first to benefit from the supply shortage of High Bandwidth Memory (HBM), with industry leaders like Micron Technology seeing their stock prices nearly double. Competitors like AMD seized market share during Nvidia's sensitive delivery window, driving up the penetration rate of the MI500X chip and significantly outperforming Nvidia in terms of first-half gains. Upstream and downstream companies responsible for wafer fabrication and advanced packaging also diverted a large amount of capital.

Wall Street's investment thesis is shifting from "buy only Nvidia" to "buying into the entire semiconductor supercycle." This broad distribution of capital directly led to Nvidia being passively marginalized during the index's surge.

Previously, the market viewed Nvidia as the largest beneficiary of the AI wave, but as AI technology penetrates broader fields, other segments of the semiconductor supply chain have also begun to benefit.

Investors realized that the development of AI requires the support of the entire supply chain, rather than just the chip manufacturing segment. Consequently, capital began diffusing into subsectors like memory, packaging, and equipment, fueling a rally across the entire semiconductor sector, while Nvidia, due to its massive previous run-up, became a target for profit-taking.

The marginal slowdown in the market's core demand for AI chips is directly reflected in the changing rental prices of data center computing power.

Monitoring data shows that the hourly rental price of Nvidia's flagship B200 GPU plunged 31% from $6.11 to $4.22 in just three weeks from late May to June. This rapid decline sends a warning signal that the growth rate of global computing power supply has begun to temporarily outpace the growth rate of new AI workloads, shaking Wall Street's confidence in Nvidia's pricing power.

Investors are concerned that as supply and demand in the computing power market move toward balance, the myth of Nvidia's ultra-high gross margins may be unsustainable. Over the past two years, Nvidia secured immense pricing power due to the supply shortage of AI chips, which once pushed its gross margins to historic highs.

However, as competitors' capacity is unleashed and marginal market demand slows, this pricing power faces challenges. The decline in GPU rental prices not only reflects changes in supply and demand, but also signals that the AI chip market may gradually shift from a seller's market to a buyer's market, meaning Nvidia will need to cope with more intense price competition while maintaining its technological leadership.

Bank of America ( BAC) pointed out in its latest research report that Nvidia's significant underperformance relative to the Philadelphia Semiconductor Index constitutes a highly attractive buying opportunity, and reiterated its Buy rating and $350 price target.

Analyst Vivek Arya summarized four core market concerns regarding Nvidia's stock: gross margin pressure from high memory costs, competitive threats from custom ASIC chips, overly concentrated investor ownership, and some investors' view that the company's use of cash through vendor financing—rather than share buybacks or dividend payouts—is inefficient.

On the margin front, Arya believes the market has overestimated HBM cost pressures while underestimating Nvidia's pricing power and scale advantages.

Following the upgrade from the Blackwell to the Vera Rubin architecture, the HBM cost per rack could increase by $200,000 to $300,000. However, driven by comprehensive upgrades in computing, networking, and software, the overall pricing of a rack is expected to rise by $2 million to $3 million. Bank of America expects gross margins to remain around the mid-70% range, while supply chain commitments of up to $119 billion also provide support for margin stability.

Regarding ASIC competition, in-house chips such as Google's TPU, Amazon's Trainium, and Meta's MTIA are viewed as potential challengers. However, Arya pointed out that these products have existed for years, whereas Nvidia's GPU revenue has grown approximately 700-fold since 2015, and its sales to hyperscalers increased 115% year-over-year—nearly double the growth rate of cloud computing capital expenditures—demonstrating its strong market resilience.

In addition, Nvidia's current forward P/E ratio is only 18 times, hitting a seven-year low, representing an unjustified 30% to 35% discount relative to peer growth stocks, which highlights its valuation attractiveness.

Nvidia's current stock price weakness is essentially "growing pains" under the pressure of high valuations. The valuation overshoot from its earlier rally, temporary diversion of market funds, and marginal changes in the supply-demand dynamics of computing power have collectively caused it to underperform the semiconductor index.

However, from a long-term perspective, Nvidia's technological barriers, ecosystem advantages, and market position in the AI chip sector remain robust, and the launch of its next-generation Vera Rubin chips is expected to be a key catalyst to break the current deadlock.

For investors, the current stock price pullback may represent an excellent opportunity to build positions. The market's overreaction to short-term fluctuations has pushed Nvidia's valuation to historical lows, providing a higher margin of safety and upside potential.

Although it may still face challenges such as slowing earnings growth and intensifying competition in the short term, the long-term growth trend of the AI industry remains unchanged. As the industry leader, Nvidia is poised to continue benefiting from technological iterations and market expansion. Remaining rational amid market disagreements and focusing on the company's core value may be the key to seizing Nvidia's investment opportunities.

This content was translated using AI and reviewed for clarity. It is for informational purposes only.