Semiconductor stocks surged 102% in half a year, crushing Bitcoin: Why is global capital flooding into AI hardware in 2026?

In the first half of 2026, global asset performance showed a rare divergence. On one hand, the Philadelphia Semiconductor Index (SOX) led all major global asset classes with a 102% gain; on the other hand, the "Magnificent Seven" stocks overall fell 2%, while Bitcoin (BTC) experienced a sharp 33% correction. During the same period, the Nasdaq Composite Index rose approximately 12.8%, and the S&P 500 Index gained less than 10%.

This pattern indicates that market capital is undergoing a structural shift—moving away from the platform-based tech companies and high-volatility crypto assets that dominated the past two years, further concentrating into AI infrastructure and the semiconductor supply chain. Understanding the driving logic behind this divergence is crucial for determining asset allocation directions in the second half of 2026 and beyond.

Why the Semiconductor Sector Became the Biggest Winner of 2026

The Philadelphia Semiconductor Index doubled in the first half of the year, with a single-quarter gain of nearly 88% in Q2, marking the strongest quarterly performance since the index's inception. Micron, AMD, Google, and Intel contributed approximately 16%, 10%, 8%, and 8% of the upward momentum, respectively. Memory and chip stocks became the best-performing sector in the S&P 500 Index—SanDisk gained about 760% year-to-date, while Micron Technology, Intel, Western Digital, and Seagate Technology all more than doubled year-to-date.

Behind the collective surge of the semiconductor sector lies the superposition of multiple structural forces.

The exponential expansion of AI computing demand constitutes the core driving force. Nvidia CEO Jensen Huang, during the earnings call, described AI data center construction as "the largest infrastructure expansion in human history." Goldman Sachs estimates that global AI capital expenditures related to computing, data centers, and electricity will reach approximately $7.6 trillion from 2026 to 2031, with annual spending rising from $765 billion in 2026 to $1.64 trillion by 2031. Hyperscaler cloud vendors' AI investments by 2030 could exceed $6 trillion.

Data center construction has entered a new expansion cycle. Morgan Stanley has significantly revised its 2026 U.S. big tech capital expenditure forecast upward from $433 billion a year ago to $805 billion, with 2027 CapEx potentially reaching $1.1 trillion. The combined CapEx of Alphabet, Amazon, Microsoft, and Meta in 2026 is expected to reach $725 billion, a 77% increase from $410 billion in 2025. In the first quarter of 2026 alone, these four companies' AI infrastructure-related CapEx reached $130 billion.

The rapid increase in demand for HBM, high-performance GPUs, and advanced manufacturing processes has directly translated into revenue and profits for chip companies. Total combined capital expenditures of major Asian semiconductor manufacturers in 2026 are expected to exceed $136 billion, up about 25% from 2025. Global IDM and foundry CapEx is estimated to reach $272 billion. Japanese semiconductor equipment and materials companies continue to benefit from this expansion cycle, with foreign investors net buying Japanese stocks for eight consecutive weeks, with weekly net inflows exceeding one trillion yen.

As the most upstream hardware suppliers in the AI supply chain, semiconductor companies can "recognize revenue immediately" in the current wave of investment. This characteristic makes them the most direct and certain beneficiaries of AI capital expenditures.

Why the "Magnificent Seven" Underperformed the Broader Market

If semiconductors are the "pick-and-shovel sellers" of the AI supply chain, then most of the "Magnificent Seven" play the role of "gold prospectors"—investing heavily, but with returns not yet fully validated.

The Bloomberg "Magnificent Seven" Total Return Index fell about 5.6% in the first half of the year. Specifically, Microsoft led the decline, dropping over 22%; Meta fell 14%; Tesla fell 6%. In June alone, the market capitalization of the "Magnificent Seven" collectively evaporated about $2.3 trillion.

The core reason for this adjustment is the shift in market pricing logic from "growth narrative" to "profitability verification." Hyperscalers like Microsoft, Amazon, Meta, and Google each invest hundreds of billions of dollars annually in data centers, but when AI businesses will generate profits commensurate with those investments remains unclear. Goldman Sachs expects that hyperscaler CapEx as a percentage of operating cash flow will rise to about 100% in 2026, reflecting that these companies are reinvesting nearly all internal cash flow into AI infrastructure.

Meanwhile, market attention has shifted from the AI application layer to AI infrastructure. Tom Lee, head of research at independent research firm Fundstrat Global Advisors, noted that the market is reinterpreting the new narrative surrounding the "Magnificent Seven"—they have transformed from asset-light companies generating substantial free cash flow into companies with heavier balance sheets. When massive capital expenditures continue to consume cash flow while returns have yet to materialize, investors begin to question the valuation rationale of these platform-based tech companies.

Deutsche Bank attributes this phenomenon to four factors: positions in large-cap tech stocks had reached "extreme" levels by the end of May; the profitability conversion capability of AI investments is being questioned; rising memory chip prices are driving up data center construction costs; and the market is transitioning from valuation-driven to profitability verification stages.

Why Bitcoin Underperformed Semiconductor Stocks

Crypto assets performed even more poorly in the first half of 2026. Bitcoin fell from around $87,500 at the start of the year to below $59,000 in June, a decline of 33%. Ethereum dropped as much as 47%, and Solana fell 41%.

Bitcoin's underperformance relative to semiconductor stocks is no coincidence.

The global liquidity environment continues to pressure crypto assets. Although Fed policy expectations have changed, overall liquidity conditions have not significantly eased, putting sustained pressure on high-volatilty digital assets. Bitcoin's nature as a risk asset makes it more vulnerable amid global macroeconomic uncertainty.

The AI theme has become the main narrative in equity markets in 2026, absorbing a significant amount of potential allocation capital. Spot Bitcoin exchange-traded funds (ETFs) recorded net outflows of $5.4 billion in the first half of 2026, marking the first semi-negative performance since their launch. In May and June alone, BlackRock's IBIT accounted for $5 billion of those outflows. DWF Labs attributed the net outflows to capital shifting toward AI investments. Spot Ethereum ETFs also recorded their first negative semi-annual performance, with net outflows of $1.47 billion. From May 15 to June 3, Bitcoin ETFs saw net outflows for 13 consecutive trading days, draining $4.4 billion from the category.

Bitcoin and semiconductors represent entirely different investment logics. Semiconductors are direct beneficiaries of AI infrastructure—chip companies can recognize revenue immediately from AI capital expenditures. Bitcoin, on the other hand, more broadly reflects changes in the digital asset market and global liquidity, and has not directly benefited from the current wave of AI construction. Brian Garrett, a derivatives expert at Goldman Sachs, noted that the market views Bitcoin as a "spending-type" asset, in stark contrast to "revenue-type" semiconductor companies.

The difference in the pricing logic of the two asset classes determines their divergent paths under the same macro environment.

Will the AI Boom Revitalize the Crypto Market?

Is the divergence between semiconductors and Bitcoin a temporary structural phenomenon or the beginning of a long-term trend? It depends on the evolution of several variables.

The convergence of AI and blockchain is breeding new growth points. Tracks such as AI agents, decentralized computing, and DePIN (Decentralized Physical Infrastructure Networks) may bring new capital and attention to the crypto market in the medium term. CoinGecko data shows that Render (RNDR) rose 17% in the first half of the year, and NEAR Protocol (NEAR) gained 18%, while most mainstream cryptocurrencies fell over 30% during the same period. These two tokens, primarily focused on computing services, became relatively scarce resources in this cycle.

Improvement in macro liquidity is a key prerequisite for capital to return. When global monetary policy turns accommodative and risk appetite rises, crypto assets could see renewed inflows of allocation capital. Some analysts recently pointed out that the AI storage and semiconductor sectors have shown notable cooling, while Bitcoin rebounded from its local low to above $61k, sparking discussions on whether capital is beginning to reallocate to digital assets.

However, it is worth noting that semiconductors and crypto assets are not a simple "seesaw" relationship. Behind them lie different industrial cycles and asset pricing logics. The vitality of the semiconductor sector depends on the sustainability of AI capital expenditures and the evolution of the chip industry cycle; the crypto market's trajectory is more influenced by global liquidity, regulatory environment, and ecosystem development progress.

As of 7:00 PM Beijing time on July 7, 2026, the three major U.S. stock indices closed higher. The Dow Jones Industrial Average rose 0.29% to 53,055.91 points, breaking above 53,000 points for the first time to hit a record high; the Nasdaq Composite gained 1.12% to 26,121.16 points; the S&P 500 Index rose 0.72% to 7,537.43 points. The Philadelphia Semiconductor Index surged 4.95% in a single day, with the entire chip supply chain strengthening across the board. Bitcoin broke above $64,000, temporarily trading at $64,159. The market is at a critical inflection point for direction.

Conclusion

The divergence in global asset performance in the first half of 2026 is essentially a mirror reflection of the AI industry cycle's evolution in capital markets. The semiconductor sector's 102% gain leading the way is driven by the triple resonance of AI computing demand, data center expansion, and the chip industry cycle; the adjustment of the "Magnificent Seven" reflects the market's shift in pricing logic from "growth narrative" to "profitability verification"; Bitcoin's 33% correction reveals the profound impact of liquidity conditions and capital flows on its asset price.

The divergence among the three asset classes highlights a core question: during the large-scale investment phase of AI, which part of the supply chain should capital be allocated to? The advantage of semiconductor companies lies in "recognizing revenue immediately"; the challenge for platform-based tech companies is the "temporal mismatch between investment and returns"; and the positioning of crypto assets requires finding their own value support beyond the AI narrative.

For investors, understanding the underlying logic of this divergence is far more important than chasing short-term price movements. When AI investment transitions from "hardware first" to "application landing," capital flows may shift again. At that time, the relative performance among semiconductors, tech giants, and crypto assets will undergo a new round of repricing.

FAQ

Q: How much did the Philadelphia Semiconductor Index specifically rise in the first half of 2026?

According to Deutsche Bank data, the Philadelphia Semiconductor Index (SOX) accumulated a 102% gain in the first half of 2026, with a single-quarter gain of nearly 88% in Q2, the strongest quarterly performance since the index's inception.

Q: Why did Bitcoin plummet in the first half of 2026?

Bitcoin fell from around $87,500 to below $59,000, a decline of 33%. The main reasons include: tight global liquidity pressuring high-volatility assets; the AI theme absorbing a large amount of potential allocation capital, with spot Bitcoin ETFs seeing net outflows of $5.4 billion in the first half; and Bitcoin, as a "spending-type" asset, failing to directly benefit from the AI construction wave.

Q: What is the difference in investment logic between semiconductor stocks and Bitcoin?

Semiconductors are direct beneficiaries of AI infrastructure—chip companies can recognize revenue immediately from AI capital expenditures. Bitcoin, on the other hand, more broadly reflects changes in the digital asset market and global liquidity, and has not directly benefited from the current wave of AI construction, being viewed by the market as an asset with performance similar to "spending-type" companies.

Q: Why did the "Magnificent Seven" underperform the broader market in 2026?

The "Magnificent Seven" overall fell 2% in the first half of the year. The core reason is the shift in market pricing logic from "growth narrative" to "profitability verification." Hyperscaler capital expenditures have continued to swell, but AI business returns have yet to materialize, leading investors to question the valuation rationale of these companies.

Q: Will the AI boom drive a rebound in the crypto market in the future?

The convergence of AI and blockchain (such as AI agents, DePIN, decentralized computing, etc.) may bring new growth points to the crypto market in the medium term. However, whether capital returns depends crucially on the improvement of global liquidity and the pace of AI investment shifting from hardware to the application layer. The difference in the pricing logic of the two asset classes means they are not a simple "seesaw" relationship.