For more than three years, the so-called 'Magnificent Seven' — Nvidia, Apple, Microsoft, Alphabet, Amazon, Meta, and Tesla — were the undisputed darlings of Wall Street, powering index gains and drawing in billions from global investors. But June 2026 marked a sharp reversal.
Nvidia fell over 5%, Microsoft dropped roughly 17% — its worst monthly performance since December 2000 — while Alphabet declined nearly 6%, Amazon lost about 12%, and Meta fell around 11%. Apple and Tesla saw volatile but divergent moves: Apple hit an all-time high of $315.2 early in the month before sliding more than 10%, while Tesla dropped over 6% in the first week but recovered to end roughly flat.
Collectively, the group shed approximately $2.3 trillion (€2 trillion) in market value in a single month. The breadth of the selloff was unusual; typically one or two stocks stumble while others hold up. This time, nearly every member moved lower. The Roundhill Magnificent Seven ETF (MAGS), which tracks all seven, fell about 13% from its late May peak.
Why the sudden shift?
The MAGS ETF bled more than $700 million (€615 million) in outflows over June, its worst month since launching in 2023, according to TradingView data. For a fund that had become the simplest way to bet on the US tech boom, the reversal was striking.
Oracle, a hyperscaler not in the Magnificent Seven, crashed around 35% — its steepest monthly drop since September 1990 — after alarming investors with a surge in AI spending and debt. The fall wiped roughly $100 billion (€87.9 billion) off the fortune of co-founder Larry Ellison. The market punished the biggest AI spenders, and the numbers explain why.
The five largest hyperscalers are on track to spend more than $700 billion (€615 billion) on AI infrastructure this year. Microsoft alone is heading toward roughly $190 billion (€167 billion), according to Bank of America estimates. The bank noted that hyperscaler capital spending has jumped from about 70% of operating cash flow in 2025 to nearly 100% in 2026.
The implication is straightforward: far less capital remains for share buybacks and dividends, and an increasingly large bill must be justified with future revenue as costs climb. Memory chips that feed AI data centres have become scarce and expensive. Micron Technology reported earnings per share of $24.67 for its latest quarter, up from $1.68 a year earlier — a nearly fifteenfold jump. Prices for DRAM, the memory inside almost every device, rose as much as 98% in the first quarter alone, a surge some in the industry have dubbed 'RAMageddon'.
A quieter shift beneath the surface
While the biggest technology stocks struggled, the rest of the market continued to rise. LPL Financial chief equity strategist Jeff Buchbinder points to that trend. Excluding the Magnificent Seven, the remaining S&P 500 companies grew earnings by 17.5% in the first quarter, helped in part by semiconductor and memory producers. Buchbinder expects that figure to exceed 20.5% in the second quarter, while earnings growth for the Magnificent Seven is projected to be lower.
In other words, the other 493 companies are now growing earnings faster than the market's biggest stars, and investors have noticed. By late June, the S&P 493 — which excludes the Magnificent Seven — had climbed 13.7% for the year. In contrast, the Magnificent Seven basket was down 6.6%, while the broader S&P 500 posted a more modest 7.4% gain.
Veteran investor Ed Yardeni suggests investors are showing signs of AI fatigue, questioning whether unprecedented spending on infrastructure will ultimately generate attractive returns as cheaper open source models proliferate and AI token prices continue to decline.
For European investors, the shift has implications. Many European pension funds and sovereign wealth funds have significant exposure to US tech stocks. Meanwhile, European tech companies — from ASML in the Netherlands to SAP in Germany — may benefit if capital rotates toward more diversified holdings. The broader market's resilience also offers a reminder that the AI boom, while transformative, is not the only game in town.
The Magnificent Seven still delivered an estimated 29% earnings growth in the first quarter, and they are unlikely to lose their leadership positions anytime soon. Yet the debate has shifted. Investors are no longer asking whether AI will transform the economy. They are asking when hundreds of billions of dollars in AI investment will begin producing meaningful returns. June may have offered the first clear answer.
