On June 15, the US stock market experienced a night that will go down in history. The Dow Jones Industrial Average closed at 51,671 points, setting a new all-time high. The Nasdaq Composite soared 3.07%, marking its largest single-day gain in nearly two and a half months. But the true star of the session was the Philadelphia Semiconductor Index (SOX), which skyrocketed 5.45% in a single day to close at 14,099 points, smashing through the previous record high set on June 3.
The catalyst for this rally can be traced clearly along a geopolitical chain of events. The United States and Iran announced a framework agreement to reopen the Strait of Hormuz, with the official signing ceremony scheduled for that Friday in Switzerland. The Strait of Hormuz is one of the world’s most critical oil shipping chokepoints. The sudden prospect of restored passage immediately triggered a sharp decline in international oil prices. WTI crude futures plunged nearly 5%, briefly touching $80.25 per barrel, while Brent crude dropped more than 4% as well.
The significance of the oil price crash goes far beyond the energy market itself. For capital markets, falling energy prices directly eased the persistent anxiety caused by high inflation. As inflation expectations cooled, concerns over the Federal Reserve maintaining a tight monetary policy faded, rapidly restoring overall risk appetite. Capital flowed out of defensive assets and into growth assets—technology stocks, and especially semiconductors, became the biggest winners of this risk-on shift.
Explosive Growth Across Memory Chips: How the Supply-Demand Gap Narrative Fueled the Rally
Behind the record-setting performance of the Philadelphia Semiconductor Index, the memory chip sector stood out in particular. Western Digital surged over 16% in a single day, posting its best daily performance since January and closing at a record high. Micron Technology jumped more than 10%. Seagate Technology climbed over 9%, also hitting a record high. ARM rose more than 8%, while AMD and SanDisk gained over 6%.
The breakout in the memory sector was not merely driven by market sentiment. Morgan Stanley’s hardware research team, after conducting channel checks, noted that demand for hard disk drives (HDDs) continues to strengthen and is broadening across more segments, creating significant upward pressure on prices. Their report projects global HDD demand will grow at an annual rate of 40% to 50%, while supply will only increase by 30% to 35%. This supply-demand gap is expected to sustain shortages at least through 2028. Based on these findings, Morgan Stanley maintained its overweight ratings on Western Digital and Seagate Technology and sharply raised their price targets.
Additionally, AMD announced the acquisition of memory optimization technology firm MEXT. MEXT’s AI-driven technology can expand memory capacity without sacrificing computing power or energy efficiency. This acquisition further bolstered market optimism about advances in memory technology. As AI computing power needs grow exponentially, storage bottlenecks are becoming a key constraint on AI development. Any technology that promises to break through this bottleneck could command a premium in the capital markets.
From a 10% Plunge to Record Highs: What Does the V-Shaped Reversal Reveal About Market Sentiment?
The Philadelphia Semiconductor Index’s V-shaped reversal over just a few trading days offers a vivid snapshot of current market sentiment.
On June 5, US stocks suffered a "Black Friday." The SOX plummeted 10.26% in a single session, its steepest one-day drop since the COVID-induced panic of March 2020. Nvidia fell 6.20%, TSMC dropped 6.69%, and Micron Technology crashed 13.25%. At that time, the prevailing narrative was "the AI bubble has burst" and "tech stocks have peaked."
Yet just 10 days later, the same group of assets staged a stunning turnaround from crash to record highs. This extreme volatility highlights the core characteristics of the current AI semiconductor sector—highly crowded trades, hypersensitive sentiment, and intensely concentrated capital flows.
It’s worth noting that before the June 5 sell-off, the SOX had surged 94.86% over the 45 trading days from March 31 to June 3. Such a steep ascent meant profit-taking pressure had built up significantly. The June 5 plunge can be seen as a sharp round of profit-taking and sentiment clearing, while the swift rebound that followed indicates that the market’s medium- to long-term narrative around AI semiconductors remains fundamentally intact. Every dip continues to be seen as a buying opportunity.
80% of Fund Managers Agree: What Does the "Most Crowded Trade in History" Mean for Semiconductors?
On the same day the SOX set a new record high, Bank of America released its latest monthly fund manager survey. Covering $465 billion in assets and conducted from June 5 to 11, the report reveals the true views of professional investors on the AI semiconductor sector.
The survey’s core findings present a fascinating contradiction.
On one hand, 56% of global fund managers described the current stage of the AI cycle as a "boom," meaning the rally continues to gather momentum and more investors are entering for fear of missing out. Only 21% believe the sector has reached the "euphoria" stage of extreme valuations, and just 9% say AI is in the "profit-taking" phase. In other words, most professionals see further upside for AI.
On the other hand, a staggering 80% of respondents said that "buying and holding global semiconductor stocks" is currently the most crowded trade in the market. This figure not only topped the survey for a second straight month, but also set a new all-time record in BofA’s fund manager survey history.
A "crowded trade" is not synonymous with a bubble. It means a large amount of capital is concentrated in the same direction, so any unexpected negative shock could trigger a stampede for the exits. But from another perspective, crowding also signals extremely strong consensus—when 80% of professional investors see semiconductors as the most crowded trade, that view is already fully priced in.
Boom and Retreat: Institutional Investors Quietly Reduce Exposure
Another set of key survey data reveals a subtle gap between institutional investors’ stated sentiment and their actual actions.
Despite the strong AI narrative, investors have quietly begun defensive portfolio adjustments. Fund managers cut their overweight exposure to the tech sector from 33% to 26%, and global equities from 50% to 38%. While cash holdings remain near historic lows, they have ticked up slightly.
This "talking boom, acting defensive" behavior typically appears in the late stages of a market rally. Investors still believe in the long-term narrative, but are starting to prepare for potential risks. Funds are rotating out of crowded AI beneficiaries and into defensive sectors like financials and telecoms, reflecting the market’s delicate balance between chasing growth and managing downside risk.
Notably, 12% of respondents also identified going long the "Magnificent Seven" tech stocks as a crowded trade. This suggests crowding is not limited to semiconductors but is spreading across the entire tech sector.
SpaceX’s $2.52 Trillion Valuation: How a Giant IPO Reshapes the AI Bubble Narrative
If the SOX’s record high reflects the current valuation levels of existing AI assets, SpaceX’s IPO adds a whole new dimension to the AI bubble debate.
Following its historic IPO on Nasdaq last week, SpaceX shares soared another 19.6% on June 15 to close at $192.50, pushing its total market cap to $2.52 trillion. Compared to its IPO price of $135, the stock has gained over 42% in just two trading days. SpaceX now ranks sixth among global public companies by market capitalization, trailing only Nvidia, Alphabet, Apple, Microsoft, and Amazon. Underwriters exercised the "greenshoe" overallotment option that day, bringing total IPO proceeds to $85.7 billion—a new global record.
However, there’s a significant disconnect between SpaceX’s financials and its valuation. The company’s full-year 2025 revenue is $18.7 billion, but it posted a net loss of $4.94 billion. In Q1 2026, revenue was about $4.7 billion, with a net loss of $4.3 billion. Although the Starlink business is profitable—contributing $11.39 billion in revenue and $4.42 billion in operating profit in 2025, for a 39% margin—massive investments in AI infrastructure continue to weigh on overall profitability.
A company that has yet to turn a profit commanding a $2.52 trillion valuation has become one of the strongest arguments in the AI bubble debate. Independent research firm CFRA has issued a "sell" rating on SpaceX and slashed its price target by more than 40%. Many analysts warn that unless SpaceX’s financials improve significantly within a certain timeframe, the foundation supporting its current valuation will be tested.
Concentration Surpassing Historical Bubbles: The AI Sector’s Scale Warning Line
BofA Chief Strategist Michael Hartnett has issued a strong warning: if you combine newly listed or soon-to-be-listed tech giants like SpaceX and OpenAI with existing AI leaders, the AI sector’s market concentration will reach about 48%.
What does this number mean? A 48% concentration surpasses the levels seen during the "Roaring Twenties" of the 1920s, the "Nifty Fifty" of the 1970s, the Japanese stock market bubble of the 1980s, and the TMT bubble of the 1990s. The only historical record it hasn’t surpassed is the 63% peak during the railroad bubble of the 1880s.
While concentration isn’t a direct measure of risk, it is a critical warning indicator. When a handful of sectors account for such a large share of the market, any negative shock to that sector can ripple across the entire market through sheer weight. This explains why institutional investors, while remaining optimistic, have started to tactically reduce their exposure to tech stocks.
Conclusion
The US stock market’s performance on June 15, 2026, was the result of multiple factors converging: easing geopolitical risk triggered a plunge in oil prices, cooling inflation expectations unlocked risk appetite, the memory chip supply-demand gap narrative provided fundamental support, and SpaceX’s massive IPO further ignited market sentiment. After a 10% plunge on June 5, the SOX staged a rapid V-shaped rebound to record highs, vividly illustrating the market’s current extreme volatility.
However, the contradictory signals from the BofA fund manager survey cannot be ignored—56% of respondents believe AI is in a "boom phase," but 80% say semiconductors are now the "most crowded trade in history." Institutional investors are quietly reducing positions, with tech sector overweights dropping from 33% to 26%. The AI sector’s market concentration is now around 48%, surpassing several historical bubble periods.
The line between "boom" and "euphoria" may only become clear in hindsight. What is certain is that the current AI semiconductor sector is in a phase of highly concentrated consensus, highly concentrated positions, and highly expanded valuations—a structure that contains both continued upside momentum and significant underlying fragility.
FAQ
Q: How did the Philadelphia Semiconductor Index perform on June 15, 2026?
A: The Philadelphia Semiconductor Index (SOX) surged 5.45% that day, closing at 14,099 points and breaking its previous record high. On June 5, the index had plunged 10.26% in a single day, marking its largest drop in nearly six years.
Q: What is the key data from the BofA fund manager survey regarding crowded semiconductor trades?
A: The survey found that 80% of fund managers believe "buying and holding global semiconductor stocks" is currently the most crowded trade in the market, the highest reading in the survey’s history. Meanwhile, 56% of respondents think the AI cycle is in a "boom" phase, while only 21% believe it has reached "euphoria."
Q: How has SpaceX performed in terms of market cap and stock price since going public?
A: SpaceX went public on Nasdaq at $135 per share on June 12. In just two trading days, the stock gained over 42%. On June 15, it closed at $192.50, giving it a total market capitalization of $2.52 trillion, ranking sixth among global public companies.
Q: What is the current level of market concentration in the AI sector?
A: According to BofA’s chief strategist, if you combine SpaceX, OpenAI, and other companies with existing AI leaders, the AI sector’s market concentration is about 48%. This surpasses the concentration levels of the "Roaring Twenties" and the 1990s TMT bubble.
Q: How have institutional investors recently adjusted their positions in the tech sector?
A: According to BofA’s survey, fund managers have reduced their tech sector overweight from 33% to 26%, and their global equities overweight from 50% to 38%, indicating that some institutions are making defensive adjustments in response to sector concentration risks.
