In the first half of 2026, global asset performance showed a rare divergence. On one side, the Philadelphia Semiconductor Index (SOX) surged 102%, leading all major asset classes worldwide. Meanwhile, the "Magnificent Seven" US tech giants collectively fell 2%, and Bitcoin (BTC) saw a sharp 33% correction. During the same period, the Nasdaq Composite rose about 12.8%, while the S&P 500 gained less than 10%.
This landscape signals a structural migration of capital—shifting away from platform-based tech companies and highly volatile crypto assets that dominated the past two years, and concentrating further on AI infrastructure and the semiconductor supply chain. Understanding the drivers behind this divergence is crucial for determining asset allocation strategies for the second half of 2026 and beyond.
Why Semiconductors Became the Biggest Winners of 2026
The Philadelphia Semiconductor Index doubled in the first half of the year, with a nearly 88% gain in Q2 alone—the strongest single-quarter performance since its inception. Four companies—Micron, AMD, Google, and Intel—contributed approximately 16%, 10%, 8%, and 8% of the index’s momentum, respectively. Memory and chip stocks emerged as the top-performing sector within the S&P 500: SanDisk soared roughly 760% year-to-date, while Micron Technology, Intel, Western Digital, and Seagate Technology each more than doubled.
The explosive growth in semiconductors is fueled by multiple structural forces.
Exponential expansion in AI computing demand is the core driver. Nvidia CEO Jensen Huang described AI data center construction as "the largest infrastructure expansion in human history" during an earnings call. Goldman Sachs estimates that between 2026 and 2031, global AI capital expenditures focused on computing, data centers, and electricity will reach about $7.6 trillion, with annual spending rising from $765 billion in 2026 to $1.64 trillion in 2031. Hyperscale cloud providers could invest over $6 trillion in AI by 2030.
Data center construction has entered a new expansion cycle. Morgan Stanley sharply revised its 2026 US tech giant capex forecast from $433 billion a year ago to $805 billion, with 2027 capex expected to hit $1.1 trillion. Alphabet, Amazon, Microsoft, and Meta are projected to spend a combined $725 billion in capex in 2026, up 77% from $410 billion in 2025. In Q1 2026 alone, these four companies allocated $130 billion to AI infrastructure-related capex.
Rapid growth in demand for HBM, high-performance GPUs, and advanced process technologies is directly translating into revenue and profits for chipmakers. Major Asian semiconductor manufacturers are expected to spend a combined $136 billion in capex in 2026, up about 25% from 2025. Global IDM and foundry capex may reach $272 billion. Japanese semiconductor equipment and materials firms continue to benefit from this expansion cycle, with foreign investors net-buying Japanese stocks for eight consecutive weeks and weekly inflows exceeding one trillion yen.
As the upstream hardware suppliers in the AI value chain, semiconductor companies can "recognize revenue immediately" during this investment wave. This makes them the most direct and certain beneficiaries of AI capital expenditure.
Why the "Magnificent Seven" Underperformed
If semiconductors are the "pick-and-shovel" sellers in the AI gold rush, most of the "Magnificent Seven" play the role of "prospectors"—making huge investments, but with returns yet to be fully validated.
Bloomberg’s "Magnificent Seven" stock return index fell about 5.6% in the first half. Specifically, Microsoft led the decline, dropping over 22%; Meta fell 14%; Tesla slipped 6%. In June alone, the group’s combined market cap evaporated by approximately $2.3 trillion.
The core reason for this correction is that the market’s pricing logic shifted from "growth narrative" to "profit validation." Microsoft, Amazon, Meta, and Google—hyperscale cloud providers—are investing hundreds of billions annually in data centers, but when AI businesses will generate profits commensurate with these investments remains unclear. Goldman Sachs projects that in 2026, hyperscale cloud providers’ capex will reach about 100% of operating cash flow, indicating that almost all internal cash flow is being reinvested into AI infrastructure.
At the same time, market focus has moved from the AI application layer to AI infrastructure. Tom Lee, Head of Research at Fundstrat Global Advisors, notes that the market is reinterpreting the new narrative around the "Magnificent Seven"—they are shifting from asset-light, high free cash flow companies to firms with heavier balance sheets. As massive capex continues to consume cash flow while returns remain unrealized, investors are questioning the valuation rationale of these platform-based tech companies.
Deutsche Bank attributes this phenomenon to four factors: large tech stock holdings reached "extreme" levels by the end of May; doubts about the profitability conversion of AI investments; rising memory chip prices pushing up data center construction costs; and the market transitioning from valuation-driven to profit-validation phases.
Why Bitcoin Lagged Semiconductor Stocks
Crypto assets performed even worse in the first half of 2026. Bitcoin dropped from about $87,500 at the start of the year to below $59,000 in June—a 33% decline. Ethereum fell 47%, and Solana plunged 41%.
Bitcoin’s underperformance relative to semiconductor stocks is no accident.
The global liquidity environment continues to pressure crypto assets. Despite shifts in Federal Reserve policy expectations, overall liquidity conditions have not significantly eased, which continues to suppress highly volatile digital assets. Bitcoin’s risk asset nature makes it especially vulnerable to global macro uncertainty.
AI emerged as the dominant theme in equity markets for 2026, siphoning off substantial potential allocation capital. Spot Bitcoin ETFs recorded a net outflow of $5.4 billion in the first half—the first negative half-year since launch. In May and June alone, BlackRock’s IBIT accounted for $5 billion of the outflow. DWF Labs attributed the net outflow to capital shifting toward AI investments. Spot Ethereum ETFs also saw their first negative half-year, with net outflows of $1.47 billion. From May 15 to June 3, Bitcoin ETFs experienced net outflows for 13 consecutive trading days, draining $4.4 billion from the category.
Bitcoin and semiconductors represent fundamentally different investment logics. Semiconductors are direct beneficiaries of AI infrastructure—chipmakers can recognize revenue immediately from AI capex. Bitcoin, on the other hand, mainly reflects the dynamics of the digital asset market and global liquidity shifts, and has not benefited directly from the current AI investment wave. Goldman Sachs derivatives specialist Brian Garrett pointed out that Bitcoin is viewed as a "spending-type" asset by the market, in stark contrast to the "income-type" semiconductor companies.
The differences in pricing logic between these asset classes dictate their divergent paths under the same macro environment.
Will the AI Boom Revive the Crypto Market?
Is the divergence between semiconductors and Bitcoin a temporary structural phenomenon or the start of a long-term trend? That depends on the evolution of several variables.
The integration of AI and blockchain is nurturing new growth drivers. The development of AI Agents, decentralized computing, and DePIN (Decentralized Physical Infrastructure Networks) could bring new capital and attention to the crypto market in the medium term. According to CoinGecko, Render (RNDR) rose 17% in the first half, NEAR Protocol (NEAR) gained 18%, while most major tokens fell over 30% during the same period. These two tokens focus on power services, becoming relatively scarce resources in this cycle.
Improvement in macro liquidity is a key prerequisite for capital to flow back. When global monetary policy turns dovish and risk appetite rises, crypto assets may see renewed allocation inflows. Some analysts recently noted that the AI storage and semiconductor sectors have cooled noticeably, while Bitcoin rebounded from its local low to above $61,000, sparking debate about whether capital is starting to rotate back into digital assets.
However, it’s important to note that semiconductors and crypto assets are not simply in a "see-saw" relationship. They are driven by different industry cycles and asset pricing logics. Semiconductor sector momentum depends on the sustainability of AI capex and the evolution of the chip industry cycle, while the crypto market is more influenced by global liquidity, regulatory conditions, and ecosystem development progress.
As of July 7, 2026, UTC, all three major US stock indices closed higher: the Dow rose 0.29% to 53,055.91, breaking above 53,000 for the first time and setting a new record; the Nasdaq gained 1.12% to 26,121.16; the S&P 500 climbed 0.72% to 7,537.43. The Philadelphia Semiconductor Index jumped 4.95% in a single day, with strength across the entire chip supply chain. Bitcoin broke above $64,000, last quoted at $64,159. The market is at a critical juncture.
Conclusion
The divergence in global asset performance in the first half of 2026 essentially mirrors the evolution of the AI industry cycle in capital markets. The semiconductor sector led with a 102% gain, driven by the triple resonance of AI computing demand, data center expansion, and the chip industry cycle. The adjustment in the "Magnificent Seven" reflects a shift in market pricing logic from "growth narrative" to "profit validation." Bitcoin’s 33% correction underscores the profound impact of liquidity conditions and capital flows on its price.
This asset class divergence highlights a central question: In the era of large-scale AI investment, which segment of the value chain should capital be allocated to? Semiconductor companies have the advantage of "immediate revenue recognition," platform-based tech giants face the challenge of "timing mismatch between investment and returns," while crypto assets must find their value proposition outside the AI narrative.
For investors, understanding the underlying logic behind this divergence is far more important than chasing short-term price swings. As AI investment moves from "hardware first" to "application deployment," capital flows may shift again. At that point, 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 rise in the first half of 2026?
According to Deutsche Bank, the Philadelphia Semiconductor Index (SOX) rose 102% in the first half of 2026, with a nearly 88% gain in Q2—the strongest single-quarter performance in the index’s history.
Q: Why did Bitcoin plunge in the first half of 2026?
Bitcoin dropped from about $87,500 to below $59,000, a 33% decline. The main reasons include tightening global liquidity pressuring high-volatility assets; the AI theme attracting substantial allocation capital, with spot Bitcoin ETFs seeing $5.4 billion in net outflows in the first half; and Bitcoin, as a "spending-type" asset, failing to benefit directly from the AI investment wave.
Q: What’s the difference between the investment logic for semiconductor stocks and Bitcoin?
Semiconductors are direct beneficiaries of AI infrastructure, with chipmakers able to recognize revenue immediately from AI capex. Bitcoin mainly reflects the digital asset market and global liquidity shifts, and has not benefited directly from the current AI investment wave. The market views Bitcoin as similar to "spending-type" companies.
Q: Why did the "Magnificent Seven" underperform the broader market in 2026?
The "Magnificent Seven" collectively fell 2% in the first half. The core reason is the market’s pricing logic shifting from "growth narrative" to "profit validation." Hyperscale cloud providers’ capex continues to balloon, but AI business returns remain unrealized, prompting investors to question their valuation rationale.
Q: Will the AI boom drive a rebound in the crypto market in the future?
The integration of AI and blockchain (such as AI Agents, DePIN, decentralized computing, etc.) could bring new growth drivers to the crypto market in the medium term. However, whether capital flows back depends on improvements in global liquidity and the pace of AI investment moving from hardware to applications. Differences in asset pricing logic mean the two are not simply in a "see-saw" relationship.
