Intel’s stock just had the kind of day that forces every chip investor to recheck their thesis. Shares of the legacy semiconductor giant fell sharply on Monday, and the rest of the sector followed it lower. The PHLX Semiconductor Index, tracked by the iShares Semiconductor ETF (NASDAQ: SOXX), dropped 3.2 percent. AMD fell 2.3 percent. Even the names that have led the AI rally for two years are now showing strain. NVIDIA held up better than peers but the broader sector painted a picture of what analysts are calling “buyer exhaustion” combined with renewed concern about whether hyperscaler data center spending plans will hold through the rest of the year.
The headline-driving story, according to a MarketWatch report, was a combination of factors that hit Intel and the broader chip complex at the same time. Inflation data raised questions about how the Federal Reserve will respond, traders started taking profits in names that had run hard since February, and analysts began openly worrying that the trillion-dollar AI capex pipeline announced by Microsoft, Meta, Alphabet, and Amazon at the start of the year may not translate into the kind of equipment orders the chip sector is currently pricing in.
Why Intel Took the Worst of It
Intel’s specific problem is that its turnaround story is now four months into the make-or-break phase. CEO Lip-Bu Tan has been in the corner office since April 2025, and the stock has been on a wild ride ever since. According to a CTech analysis published earlier this year after Intel’s Q4 2025 earnings, the company rallied more than 118 percent from mid-September through January 22 before getting hit with a 17 percent single-day decline on disappointing Q1 guidance. That January selloff was Intel’s worst day since August 2024.
The bull case that drove Intel’s rally was straightforward. Tan brought in a new engineering culture, cut about 22,000 employees, and turned what looked like a confrontation with President Trump into a $10 billion federal investment that made the U.S. government a major Intel shareholder. SoftBank put $2 billion into the stock. NVIDIA invested $5 billion and announced a collaboration to integrate Intel CPUs with NVIDIA AI chips for next-generation data center systems. Reports emerged that Intel’s Arizona facility was packaging chips for NVIDIA, Tesla, and Microsoft. Production capacity for Intel’s CPUs was reportedly sold out for 2026.
That last point is the one that matters for understanding Monday’s action. Intel is supply-constrained, not demand-constrained. The company exited Q4 2025 with $13.7 billion in revenue, down 4 percent year over year, but data center revenue was up 9 percent and foundry revenue was up 4 percent. The decline came from PC sales, which is the segment most exposed to consumer weakness and most disconnected from the AI buildout. The guidance for Q1 of $11.7 billion to $12.7 billion came in below Wall Street’s $12.6 billion estimate, and Intel projected zero adjusted profit versus analyst expectations of 6 cents per share.
Tan’s explanation was that the issue is supply, not demand. Intel relied heavily on inventory in Q4 to satisfy demand, and that inventory is now exhausted. The transition to the new 18A manufacturing process is taking longer than the company expected, and Intel is spending billions on equipment to relieve the bottleneck. Tan called it a “bottleneck of success.” Investors called it disappointing, and the stock has been working off that disappointment for four months.
Monday’s selloff was the next chapter. Intel CFO David Zinsner has confirmed that the buffer inventory is depleted. Customers are starting to look at second-source options. AMD is taking server CPU share in some markets. The longer Intel stays supply-constrained, the more revenue it leaves on the table and the more credibility it loses with the customers it just won back.
The Buyer Exhaustion Story
What is happening more broadly is that the chip sector has run very hard for a very long time. NVIDIA crossed $4.4 trillion in market cap earlier this year. The iShares Semiconductor ETF has more than doubled from its August 2023 low. Every cycle has a peak, and analysts are now openly asking whether this one is closer to a peak than a midpoint.
The “buyer exhaustion” framing is not about absolute demand. Demand for AI chips is still strong. Hyperscalers are still spending. NVIDIA’s data center business is still growing at triple-digit rates year over year. The question is about the marginal buyer. After two years of relentless inflows, who is left to bid the next round of capacity up? The valuation is now stretched enough that the marginal hedge fund manager has to find a new reason to add to existing positions, and “more of the same” is starting to feel insufficient.
A $1 trillion selloff in chip stocks hit the sector in February 2026 when hyperscaler AI spending plans first triggered concerns about return on investment. The market recovered through the spring as earnings reports showed that the capex was actually translating into revenue. But the February episode set a template. Whenever there is doubt about whether the dollars hyperscalers commit to spend will actually be spent, the chip sector takes the hit. Monday looked a lot like that template repeating.
What the Inflation Data Did
The other factor in Monday’s selloff was the April CPI report. Inflation came in hotter than expected on certain components, which raised the probability that the Federal Reserve will keep rates higher for longer. That matters for chip stocks for two reasons.
First, the entire AI capex thesis depends on hyperscalers being able to fund the buildout from their cash flow and from cheap debt. If rates stay elevated, the cost of capital for the next round of investment goes up. Hyperscalers have generally been self-funding their AI buildouts from operating cash flow, but the smaller cloud providers, the startups, and the GPU-as-a-service businesses are more sensitive to rate moves.
Second, persistent inflation raises the risk of a consumer slowdown that would hit the parts of the chip business that are not AI-related. PCs, smartphones, automotive chips, and industrial semiconductors are all sensitive to broader economic conditions. Intel’s PC business already declined 7 percent year over year in Q4. If consumer demand weakens further, that pressure spreads to other segments. The market is pricing in some of that risk now.
How This Compares to Past Chip Selloffs
The 2018 chip selloff is the most direct historical comparison. That episode lasted from June through December 2018 and saw the Philadelphia Semiconductor Index drop more than 25 percent before bottoming. The 2022 selloff was even sharper, with SOXX falling more than 40 percent peak to trough as memory prices collapsed and the post-COVID demand pull-forward unwound.
Both selloffs were ultimately buying opportunities. The 2018 bottom marked the start of a multi-year rally. The 2022 bottom set the stage for the AI-driven boom that has dominated the last two years. The pattern in both cases was that the fundamental demand story remained intact even as sentiment collapsed, and investors who bought the panic were rewarded.
Whether this episode follows the same pattern depends on whether the fundamental demand picture holds. The signals are mixed. Hyperscaler capex commitments remain at record levels. NVIDIA’s order book is still strong. The data center segment of Intel, AMD, and Broadcom all grew through Q1. But the marginal buyer is hesitant, and the valuation cushion is thinner than it has been in any cycle since the dot-com era. The setup is different enough from 2018 and 2022 that the playbook may not apply cleanly.
What Investors Should Watch Next
The next major catalysts are the second-quarter earnings reports from NVIDIA, Broadcom, and the major hyperscalers. Those will tell the market whether the AI capex pipeline is actually being deployed as promised. NVIDIA’s data center revenue is the single most important number in the sector right now. If it grows in line with expectations, the buyer exhaustion narrative loses force. If it disappoints, the selloff continues.
For Intel specifically, the test is whether the 18A production ramp delivers on schedule. Tan has said supply and profitability should improve as early as Q2 2026. That guidance is what the market is now skeptical of. If Intel hits its production targets, the stock has room to rally. If it slips again, the credibility damage from another miss would be hard to recover from.
Investors who want to play chip selloffs without the single-stock risk can use the SOXX or the VanEck Semiconductor ETF (NASDAQ: SMH) as broad-based vehicles. The ETFs include the full sector and reduce the impact of any single company’s execution issues. For those who want pure-play AI exposure, NVIDIA and Broadcom remain the cleanest plays. AMD has been gaining share on Intel but is also more exposed to consumer end markets. The best AI stocks to buy in 2026 and the AI startup valuation debate are both relevant reading for investors trying to figure out where this cycle goes from here.
The Bigger Picture for Tech
Monday’s chip selloff is a reminder that the AI trade is not invulnerable. Two years of nearly uninterrupted gains have built up a lot of speculative positioning, and that positioning can unwind quickly when the narrative shifts. The fundamental demand for AI compute remains very strong. The companies building the infrastructure are still spending. The pace of model improvement is still accelerating. But the stocks have already priced in a lot of that, and the next leg up will require either an acceleration of the existing trends or the arrival of new ones.
Israel’s Unit 8200 cybersecurity empire and the broader Israeli AI talent pipeline are part of that next-leg story. Israeli semiconductor design firms, AI software companies, and chip-adjacent startups are increasingly woven into the global AI supply chain. Intel itself has significant operations in Israel, with the Kiryat Gat fab being one of the company’s most important production facilities. The geopolitical security of those assets is one of the reasons Intel’s Israeli operations are strategically valuable to the U.S. government and to the company’s recovery story.
Monday was a bad day for chip investors. The bigger question is whether it was a one-off shakeout or the start of something deeper. The next round of earnings will tell. For now, the buyer exhaustion thesis is winning the tactical argument and the structural AI bull case is still intact. The smart money is watching both.
Frequently Asked Questions
Why did Intel stock fall today?
Intel led the chip sector lower on Monday after a combination of buyer exhaustion concerns, hotter-than-expected April inflation data that raised questions about Fed policy, and renewed worries about whether hyperscaler AI capex commitments will hold through the rest of the year. SOXX fell 3.2 percent, AMD dropped 2.3 percent, and Intel was hit hardest at down roughly 6.8 percent.
What is "buyer exhaustion" in chip stocks?
Buyer exhaustion is the idea that after a long rally, the marginal buyer becomes harder to find. The fundamental demand for AI chips is still strong, but valuations are stretched and positioning is crowded. Hedge funds and large institutional investors need new reasons to add to existing positions, and “more of the same” is starting to feel insufficient at current valuations.
Is Intel's turnaround failing?
Not necessarily. Intel is supply-constrained rather than demand-constrained, and CEO Lip-Bu Tan has called the situation a “bottleneck of success.” The transition to the new 18A manufacturing process is taking longer than expected, but the company has secured a $10 billion U.S. government investment, a $5 billion NVIDIA investment, and a $2 billion SoftBank stake. The next test is whether 18A production ramps as promised in Q2 2026.
How does this selloff compare to past chip downturns?
The 2018 chip selloff saw SOXX drop more than 25 percent before bottoming, and the 2022 selloff was even sharper at more than 40 percent peak to trough. Both ended up being buying opportunities. The current situation differs in that valuations are higher and the AI demand story is more concentrated in a small number of hyperscalers. The playbook from past cycles may not apply cleanly.
What should investors do during a chip sector selloff?
Long-term investors with a multi-year horizon historically have been rewarded by buying chip stocks during corrections. The SOXX or VanEck Semiconductor ETF (SMH) provide diversified exposure. Pure-play AI investors typically focus on NVIDIA and Broadcom. The key catalysts to watch are Q2 earnings reports from NVIDIA, Broadcom, and the major hyperscalers, which will reveal whether the AI capex pipeline is actually being deployed.
Why does Israel matter for the chip sector?
Israel is a major hub for semiconductor design, AI software, and chip-adjacent startups. Intel’s Kiryat Gat fab is one of the company’s most important production facilities. The Israeli AI and cybersecurity ecosystem, anchored by Unit 8200 alumni, is woven into the global AI supply chain in ways that make the country strategically valuable to the U.S. semiconductor industry.
