NVIDIA Falls Below $5 Trillion Market Cap as Global Tech Sell-Off Intensifies

A wave of risk aversion swept through global technology markets on June 10 and 11, wiping out $620 billion in combined market capitalization across the Nasdaq Composite and sending NVIDIA shares tumbling 8.7% over two trading sessions. The Santa Clara-based chipmaker, which had briefly touched a $5.3 trillion valuation in late May, closed Wednesday at $4.87 trillion, its lowest level since March 2026. The sell-off was triggered by a confluence of factors: disappointing quarterly guidance from Oracle regarding its AI infrastructure spending, hawkish signals from the Federal Reserve, and growing investor anxiety about whether the hundreds of billions flowing into AI data centers will generate commensurate returns. Wall Street is now engaged in its most heated debate since the dot-com era: Is the AI investment cycle a bubble waiting to burst, or a structural transformation that merely needs time to prove its worth?

The Oracle Warning That Started the Slide

The immediate catalyst for the sell-off was Oracle's fiscal Q4 earnings call on June 10, delivered by CEO Safra Catz. While the company beat revenue expectations with $15.2 billion in quarterly sales, Catz cautioned that "the pace of AI infrastructure deployment is normalizing after an unsustainable sprint." She disclosed that Oracle's capital expenditure on AI data centers would decline from $8.4 billion in fiscal 2026 to an estimated $6.1 billion in fiscal 2027, a 27% reduction that caught investors off guard.

Oracle shares dropped 12.3% in after-hours trading, their steepest single-session decline since September 2022. The ripple effect was immediate. NVIDIA fell 4.1% in extended trading that evening, and by the time markets opened on June 11, the selling had metastasized across the semiconductor sector. AMD declined 6.8%, Broadcom shed 5.4%, and even TSMC, the world's most valuable chip manufacturer, fell 4.9%.

"Oracle was the first major customer to publicly admit that the AI capex cycle has a ceiling," said Morgan Stanley semiconductor analyst Joseph Moore. "When a company that has been one of the most aggressive builders of AI infrastructure signals a pullback, the market listens."

NVIDIA's Valuation Under the Microscope

At $4.87 trillion, NVIDIA remains the world's second most valuable publicly traded company, trailing only Apple. But the stock's forward price-to-earnings ratio has compressed from 52x in May to 44x as of Wednesday's close, a valuation multiple that, while still premium, reflects growing skepticism about the durability of NVIDIA's growth trajectory.

The company's most recent quarterly results, reported on May 28, showed data center revenue of $39.1 billion, up 69% year-over-year, and total revenue of $44.8 billion. Gross margins expanded to 78.2%, driven by the mix shift toward the company's high-end Blackwell Ultra GPUs, which command average selling prices exceeding $35,000 per unit. Those are extraordinary numbers by any historical standard, but they represent a deceleration from the triple-digit growth rates that defined NVIDIA's fiscal 2025.

"The question is no longer whether NVIDIA is growing. It is whether the growth rate justifies a nearly $5 trillion price tag," wrote Goldman Sachs analyst Toshiya Hari in a June 11 note to clients. Hari maintained his Buy rating and $165 price target, arguing that the current sell-off represents "a buying opportunity for patient investors." However, he acknowledged that "the margin of error has narrowed considerably."

The AI ROI Debate Intensifies

At the heart of the market's anxiety is a deceptively simple question: Is artificial intelligence actually making companies money, or is it merely consuming capital? The answer, like most things in a nascent technology cycle, depends on where you look.

A June 8 report from the Boston Consulting Group (BCG) provides some of the most comprehensive data to date. The firm surveyed 1,400 enterprises across 12 industries that have deployed AI at scale. Among companies that invested more than $50 million in AI infrastructure, 61% reported positive ROI within 18 months, up from 34% in the comparable 2024 survey. However, the distribution is heavily skewed. Technology and financial services companies account for 72% of positive-ROI cases, while manufacturing, healthcare, and retail lag significantly behind.

"The AI revolution is real, but it is not evenly distributed," said BCG managing director Vladimir Lukic. "If you are a software company, AI is a force multiplier. If you are a hospital system or a factory, the path to ROI is longer, more expensive, and more uncertain."

This uneven distribution creates a structural problem for NVIDIA. The company's revenue is increasingly dependent on a small number of hyperscale customers: Microsoft, Google, Amazon, and Meta collectively account for an estimated 58% of data center GPU purchases. If any of these companies follow Oracle's lead and moderate their spending, the impact on NVIDIA's revenue would be substantial.

Analysts Divided: Bubble or Correction?

The sell-off has sharpened a long-simmering debate among market strategists. On one side are bulls who view the current correction as a healthy repricing within a structural growth story. On the other are bears who warn of parallels to the dot-com bubble of 2000.

Morgan Stanley's Moore falls firmly in the bullish camp. "We have seen this movie before, but with a different ending," he told HotTrends. "In 1999, companies were being valued on eyeballs and clicks with no revenue. NVIDIA is generating $180 billion in annual revenue with 78% gross margins. The comparison to the dot-com era is intellectually lazy."

Not everyone agrees. Michael Burry, the investor famous for predicting the 2008 housing crisis, posted on social media platform X on June 11: "The four most dangerous words in investing remain 'this time is different.' AI capex is the new fiber-optic buildout. The parallels are unmistakable." Burry's Scion Asset Management disclosed in its most recent 13F filing that it holds put options on both NVIDIA and AMD worth a combined $890 million.

A more nuanced view comes from Deutsche Bank's chief equity strategist, Binky Chadha. "This is neither a bubble nor a blip," Chadha wrote in a June 12 research note. "We are in the painful middle phase of a technology investment cycle, where early adopters have captured the easy gains and the broader market is waiting for proof that AI can transform industries beyond technology and finance. History suggests this phase lasts 18 to 24 months."

What Comes Next for NVIDIA and the AI Trade

Despite the sell-off, NVIDIA's fundamental position remains strong. The company controls an estimated 88% of the data center AI accelerator market, and its CUDA software ecosystem creates a moat that competitors have failed to breach. The Blackwell Ultra GPU, which began volume shipments in March 2026, is sold out through the end of the year, with a backlog estimated at $45 billion.

Jensen Huang, NVIDIA's co-founder and CEO, addressed the sell-off indirectly during a June 12 appearance at COMPUTEX in Taipei. "The demand for AI compute is not slowing down," he said. "What is changing is that customers are becoming more sophisticated about how they deploy it. That is a sign of maturity, not weakness."

The next major test comes on August 27, when NVIDIA reports its fiscal Q2 2027 results. Analysts expect revenue of $47.2 billion, representing 54% year-over-year growth, with data center revenue accounting for 87% of the total. Any deviation from these expectations, particularly on the forward guidance, could either restore confidence or deepen the sell-off.

In the meantime, the broader question of AI's economic viability will continue to dominate the conversation. As one veteran hedge fund manager, speaking on condition of anonymity, told HotTrends: "Everyone agrees AI is the future. The fight is about how much the future is worth today, and whether the people building it can afford to wait."