Nvidia, Micron trade at decade-low valuations after $1 trillion selloff

Semiconductor stocks have lost more than $1 trillion in market value as investors question AI spending returns, even though demand for advanced GPUs and high-bandwidth memory remains exceptionally strong.

The selling accelerated after SK Hynix reported record quarterly results that still failed to impress Wall Street, highlighting how elevated expectations have become for AI-related chipmakers. "The fundamental AI investment thesis appears largely unchanged," Zacks Investment Research analysts wrote, noting that cloud hyperscalers continue committing hundreds of billions of dollars toward AI infrastructure.

Nvidia now trades at roughly 21 times forward earnings, near its lowest valuation in a decade and on par with the S&P 500, despite commanding the AI accelerator market. The company's fiscal 2027 and fiscal 2028 EPS estimates have risen more than 4 percent in the past two months, with annual earnings projected to climb 90 percent this year and another 38 percent next year to $12.57 per share. Revenue is projected to surpass $540 billion by fiscal 2028, more than doubling from last year's record of $215.94 billion.

Micron, trading at just 10 times forward earnings, has seen analysts raise current-quarter EPS forecasts by 41 percent and next-quarter estimates by nearly 48 percent in the past 60 days. Samsung said memory shortages could persist through 2028, reinforcing expectations that AI-driven memory demand will remain strong for years. For investors with a longer horizon, the correction may represent an opportunity to accumulate shares of industry leaders rather than a reason to abandon the AI theme.

Why the Selloff May Be Overdone

The broader chip correction appears driven more by valuation concerns and profit taking than deteriorating business fundamentals. Several factors combined to pressure chip stocks: disappointing investor reactions to SK Hynix's earnings despite record profits, questions about the pace of AI monetization, concerns over hyperscaler capital spending, and increased competition from Chinese memory producers.

Investors have shifted from asking whether AI demand is real to asking how quickly enormous infrastructure investments will translate into sustainable profits. That distinction matters. The AI revolution has not stalled; expectations have simply become extraordinarily high after semiconductor stocks generated massive gains over the past year. Corrections of this magnitude are not uncommon following such powerful rallies, especially in one of the market's fastest-growing industries.

Nvidia and Micron: Two Sides of the AI Trade

Nvidia's GPUs continue to power the overwhelming majority of generative AI training and inference workloads across leading cloud providers, while its CUDA software stack has created a competitive moat that rivals have struggled to overcome. Microsoft, Amazon, Meta Platforms, Alphabet, and Oracle continue investing aggressively in AI infrastructure, alongside numerous sovereign AI initiatives.

Micron remains one of the biggest beneficiaries of booming demand for HBM, an essential component powering Nvidia's AI accelerators. Samsung recently indicated that memory shortages could persist through 2028, reinforcing expectations that AI-driven memory demand should remain strong for years. Much of the selling appears tied to investor disappointment following SK Hynix's earnings report rather than any significant deterioration in Micron's own business outlook.

The earnings estimate revision trend for Micron is particularly notable. Over the past 60 days, analysts have raised current-quarter EPS forecasts by 41 percent, next-quarter estimates by nearly 48 percent, and full-year fiscal 2026 and fiscal 2027 projections by 24 percent and 53 percent, respectively. Rather than cutting estimates following the selloff, analysts have continued raising profit forecasts, reflecting confidence that AI-related memory demand and favorable pricing conditions remain intact.

Micron's annual revenue is projected to surge to $248 billion by fiscal 2027, representing more than a sixfold increase from last year's peak of $37.38 billion. EPS is expected to accelerate more than 790 percent this year and climb another 113 percent in fiscal 2027 to $157.83 per share.

AI Spending Is Still in the Early Innings

Major technology companies are committing unprecedented amounts of capital toward AI infrastructure because they view AI as a foundational technology rather than a discretionary investment. While Wall Street may periodically question whether returns will justify today's spending levels, history suggests that major technologies often experience periods of excessive optimism followed by sharp corrections before resuming their long-term growth path.

Nvidia has produced gains of more than 13,000 percent in the past decade, while Micron has surged nearly 6,000 percent. Both stocks currently carry a Zacks Rank #1 (Strong Buy), reflecting favorable earnings estimate revisions and compelling long-term growth prospects.

Volatility surrounding chip stocks is likely to remain elevated as investors reassess AI valuations and await additional evidence that massive infrastructure investments are translating into sustainable earnings growth. However, the long-term demand drivers supporting advanced GPUs and AI memory solutions remain firmly in place.

This article is for informational purposes only and does not constitute investment advice.