Key Highlights

  • AI semiconductor stocks extended Wednesday's steep losses into Thursday trading.
  • Investors are reassessing stretched valuations after months of AI-driven outperformance.
  • Meta Platforms (NASDAQ: META) plans to monetize excess AI computing capacity through a cloud business.
  • JPMorgan warns semiconductor stocks have significantly outperformed hyperscalers, creating valuation risks.
  • Markets are awaiting the latest U.S. employment report for clues on interest rate expectations.

AI Chip Stock Selloff Deepens as Investors Reassess Valuations

The broad-based selloff across semiconductor stocks continued into Thursday's premarket session as investors took profits in one of Wall Street's strongest-performing sectors. The weakness follows months of exceptional gains driven by artificial intelligence enthusiasm, with concerns emerging that valuations have moved well ahead of underlying earnings expectations.

Shares of Micron Technology (NASDAQ: MU) and Western Digital (NASDAQ: WDC) declined further in premarket trading after suffering steep losses during Wednesday's session. Coherent Corp. (NYSE: COHR), Marvell Technology (NASDAQ: MRVL), Advanced Micro Devices (NASDAQ: AMD), Intel Corporation (NASDAQ: INTC), and Microchip Technology (NASDAQ: MCHP) also traded lower, extending pressure across memory, networking, analog, and processor segments.

The weakness came after the semiconductor index suffered one of its sharpest declines in months, with several industry leaders—including Applied Materials (NASDAQ: AMAT), Lam Research (NASDAQ: LRCX) and Allegro MicroSystems (NASDAQ: ALGM)—posting double-digit percentage losses.

Why Investors Are Becoming More Cautious on AI Semiconductor Stocks

The recent correction reflects more than simple profit-taking. Institutional investors are increasingly debating whether semiconductor companies have become too expensive relative to their long-term earnings potential.

Over the past year, AI chipmakers have substantially outperformed the large cloud providers that ultimately purchase much of their hardware. While AI infrastructure spending remains robust, some analysts argue that this imbalance is unlikely to persist indefinitely.

JPMorgan strategist Nikolaos Panigirtzoglou noted that semiconductor and memory stocks have significantly outpaced hyperscale cloud companies since late last year. Such divergence, he argued, may prove difficult to sustain as AI investment gradually shifts from infrastructure buildout toward software monetization and application development.

The valuation debate has become increasingly important as many semiconductor companies now trade at multiples that assume years of elevated revenue growth and expanding profit margins.

Meta Platforms (NASDAQ: META) Adds a New Twist to the AI Infrastructure Story

While most semiconductor companies traded sharply lower, Meta Platforms (NASDAQ: META) stood out as one of Wednesday's strongest performers after reports suggested the company is exploring a commercial AI cloud business.

According to reports, Meta is considering selling access to unused AI computing capacity or offering customers hosted AI models running on its own infrastructure.

Such a move could help investors better understand the economic return on the company's enormous AI investments.

Earlier this year, Meta announced capital expenditure plans of up to $145 billion for 2026, one of the largest infrastructure spending programs among global technology companies. Much of that investment is directed toward building AI data centers and acquiring graphics processing units needed to train increasingly sophisticated large language models.

If Meta succeeds in commercializing excess computing capacity, investors may view those expenditures as revenue-generating assets rather than purely strategic investments.

What Meta's Cloud Expansion Could Mean for AI Infrastructure

Meta's potential entry into cloud infrastructure introduces another competitive dynamic within the rapidly expanding AI ecosystem.

Today, enterprise AI infrastructure is largely dominated by Amazon.com (NASDAQ: AMZN) through Amazon Web Services, Microsoft Corporation (NASDAQ: MSFT) through Azure, Alphabet Inc. (NASDAQ: GOOGL) via Google Cloud, and specialized AI cloud providers such as CoreWeave (NASDAQ: CRWV).

Adding Meta to that competitive landscape could create both opportunities and challenges.

On one hand, another major customer monetizing AI infrastructure reinforces the long-term demand for semiconductors, networking equipment, storage systems, and memory products.

On the other, increased competition among cloud providers may eventually reduce pricing power, potentially leading investors to reassess how quickly AI infrastructure spending will translate into sustainable profits across the broader ecosystem.

Macroeconomic Uncertainty Adds to Market Volatility

The semiconductor weakness also comes ahead of one of the week's most closely watched economic releases—the monthly U.S. employment report.

The labor market data could influence expectations for Federal Reserve policy during the second half of the year. Strong employment figures could reinforce expectations that interest rates remain elevated for longer, while weaker data may revive hopes for policy easing.

Growth-oriented technology sectors, particularly richly valued AI stocks, remain highly sensitive to changes in interest rate expectations because much of their valuation depends on future earnings growth.

With U.S. markets scheduled to close for the Independence Day holiday on Friday, many investors appear to be reducing risk ahead of both the economic data and the long weekend.

Wall Street Outlook

The recent correction does not necessarily indicate that the AI investment cycle has ended. Demand for high-performance computing, advanced memory, networking hardware, and AI accelerators remains exceptionally strong as technology companies continue expanding their data center footprints.

However, the latest selloff highlights that valuation discipline has returned to the market after an extended period of AI-driven optimism. Investors are increasingly distinguishing between companies benefiting from sustainable earnings growth and those whose share prices have already priced in years of flawless execution.

For semiconductor companies, future performance will likely depend less on AI enthusiasm alone and more on demonstrating durable revenue growth, expanding margins, disciplined capital allocation, and the ability to convert unprecedented infrastructure demand into long-term shareholder returns.