Intel stock falls 29% on weak Q4 guidance, an Israel fab blast and the memory crunch

"We face near-term challenges, including manufacturing supply shortages, the unprecedented memory crunch, and the resulting softness in PC demand." (kfgo.com) That's Intel's CEO, on the call that sank the stock. He wasn't wrong. The market just didn't care about anything else he said.

The deal

The guide

Intel's Q4 outlook is the whole story. For the fourth quarter of 2026, Intel expects revenues between $15 billion and $16 billion (yahoo.com). Intel said it expects adjusted earnings per share of 25 cents this quarter (cnbc.com). Analysts were expecting revenue of $17.1 billion and EPS of 43 cents, according to LSEG (cnbc.com).

The miss is bigger than the round number in the headline suggests. At the $15.5 billion midpoint, the guide sits roughly $1.6 billion below the $17.07 billion analysts expected (kucoin.com). Even the top of the range lands under the Street.

The reaction

Intel shares collapsed 28.6% on Friday, dragging the SOX index down 7.2% (coinotag.com). The stock fell to $65.76, wiping $146 billion in market value (kucoin.com).

The damage was ugly for the biggest holder too. The plunge stripped about $12 billion from the value of the United States government's roughly 10% stake (coinotag.com).

The irony

This came after a clean quarter. Third-quarter adjusted earnings were $0.52 per share, beating analyst estimates of $0.40 (investing.com). Management called it the eighth consecutive quarter of exceeding financial expectations (investing.com). Eight beats in a row. Then one guide erased the goodwill.

Why now

Three hits at once

Intel stacked three problems into one quarter. A memory shortage is squeezing PCs. Customers are drawing down inventory. And a fab went dark.

An explosion on October 14 shut Fab 28, Intel's plant in Kiryat Gat, Israel (tradingkey.com). Zinsner said the explosion killed two people and seriously injured one (biggo.com). The human cost comes first. The financial cost is second, but real.

Management expects the shutdown to cut about $300 million from Q4 revenue and about 2 percentage points from gross margin (tradingkey.com). The Kiryat Gat fab largely produces data center chips using Intel's 10 nanometer manufacturing process (theedgemalaysia.com). That is the product line Intel is trying to sell more of.

The memory crunch

The bigger drag is PCs. Memory shortages are increasing costs for computer manufacturers, putting pressure on system prices and making consumers and businesses more cautious (colombiaone.com). Counterpoint Research says global PC shipments fell 4% year over year in the second quarter (biggo.com).

Intel put numbers on it. Intel now expects PC units to fall about 12% this year and 10% in 2027 (tradingkey.com). The company expects PC revenue to decline in the mid-20% range in the fourth quarter, Zinsner said (theedgemalaysia.com).

My take? The Israel blast is the headline, but the PC slump is the bill. The Kiryat Gat disruption accounts for approximately two percentage points of the projected decline (colombiaone.com). The rest of the gap is structural, and it runs into next year.

What the money signals

Data center is carrying the company

Look past the stock and the business is split in two. Client computing revenue fell 5% to $8.1 billion, while data-center and AI revenue jumped 78% to $7.3 billion (biggo.com). Zinsner called it the strongest data center revenue growth in 15 years (sec.gov).

DCAI revenue is on track to finish 2026 nearly $10 billion above expectations entering the year (yahoo.com). That's real money. Servers are the growth engine.

Foundry still bleeds

Foundry is the open question. Intel Foundry revenue grew to $6.70 billion, but it posted a $1.85 billion operating loss (yahoo.com). Management itself says considerable work remains to deliver a sustainably profitable foundry business (yahoo.com).

There is progress on 18A. Intel 18A client CPUs have grown from less than 5% of revenue in Q1 to nearly one-third of client CPU revenue (yahoo.com). Yield and volume are moving the right way. Profit isn't there yet.

Cash cushion, margin cliff

Intel can absorb a bad quarter financially. It ended Q3 with approximately $54 billion in cash and short-term investments, plus $15 billion of borrowing capacity under its new extended revolvers (benzinga.com).

Margin is the problem. Intel forecast a non-GAAP gross margin of 38% for Q4, down from 44.1% in Q3 (kucoin.com). That's a steep drop for one quarter. The guidance tests the case for a stock that had more than doubled this year (biggo.com). Priced for a turnaround, then handed a setback.

The people

Lip-Bu Tan

Intel's CEO owned the miss. "I'm disappointed that we were not able to meet strong customer demand in this quarter." (kfgo.com) His pitch stays the same.

"demand for our CPUs remains robust" (sec.gov)

Lip-Bu Tan, Intel CEO

Demand is there. Supply isn't. That's a fixable problem, but not a quick one.

David Zinsner

The CFO delivered the unwelcome timeline. "We see the PC slowdown continuing at least through the first half of next year," he said (cnbc.com). He also set a bar for recovery. He said Intel does not expect a sustained recovery until memory availability improves and pricing begins to normalize (tradingkey.com).

That ties Intel's PC business to memory makers, which Intel doesn't control. Rising memory prices are a boon for memory makers but a cost burden for Intel, which buys those chips to build PC and server products (biggo.com).

The skeptics

Jim Cramer reversed course. He said he was wrong on Intel and that "They're not ready to do the things that I thought it was." (yahoo.com) Cramer isn't a primary source on semiconductors, but his flip tracks the retail mood.

The professional read is narrower. Hendi Susanto of Gabelli Funds said Kiryat Gat is one of Intel's main manufacturing sites and that the question is whether Intel can rebuild capacity fast enough to capture this wave of demand (biggo.com). That's the right question. Capacity is the bottleneck, not interest.

The backdrop

Tan inherited a capital-heavy plan. Under former CEO Pat Gelsinger, Intel poured huge sums into production facilities but struggled to secure outside customers capable of competing with TSMC (biggo.com). Tan is trying to fill those fabs. One bad quarter doesn't change that job.

The risk

Server supply is the real exposure

The comfortable story is that AI servers cover for PCs. The Israel blast complicates it. Intel expects the explosion to erase roughly $600 million in revenue over two quarters and to reduce supplies of AI server chips to cloud customers (heygotrade.com).

Intel 3 output and substrate supply both now cap data center sales (yahoo.com). If supply is the cap, demand strength doesn't help until Intel ships more.

Inventory and spillover

There's a second-order risk. Intel expects customers to reduce inventory in the fourth quarter as they adjust purchases to weaker consumption expectations (colombiaone.com). Channel destocking can drag on longer than the shortage that triggered it.

The sector felt it too. Nvidia, Broadcom, AMD, Micron, Qualcomm and Arm Holdings' US-listed shares fell between 1% and 6% in extended trading (kfgo.com). Intel named shortages of advanced substrates, memory and other critical components as obstacles to raising shipments (colombiaone.com). Those constraints aren't Intel-specific.

What can still go wrong

Execution risk remains. Intel acknowledged its ability to mitigate memory shortages remains limited (colombiaone.com). Investors should watch yield and wafer-salvage progress once Kiryat Gat resumes production, and whether soaring memory prices further squeeze PC customers (biggo.com).

Intel said most of the wafer scrapping tied to the shutdown should happen in the fourth quarter (investing.com). That gives a clean read on Q1, if nothing else breaks.

What this means

For the stock

Intel didn't break on a bad quarter. It broke on a priced-for-perfection stock meeting a guide that reset the earnings base. The Q3 numbers were fine. The next two are the problem.

I read the sell-off as partly overdone and partly earned. Overdone, because the data center business is growing at its best pace in 15 years and the balance sheet is solid. Earned, because the stock had more than doubled and the margin guide fell hard.

For developers and buyers

If you're buying hardware, plan for higher prices and slower refreshes. Intel itself says PC weakness runs through at least the first half of next year. Memory pricing sets the pace, not CPU supply.

If you run servers on Intel, ask your vendor about Granite Rapids and Clearwater Forest lead times now. Intel's stated strategic initiatives include expanding server supply and progressing on those two server roadmaps (benzinga.com). Lead times are the number to pin down. Intel says the blast reduces AI server chip supply to cloud customers, so ask your cloud provider about Intel-based capacity too.

What to watch

Three things settle this.

• Kiryat Gat restart: yields and how fast Intel scraps and replaces wafers.
• Memory pricing: Zinsner ties any sustained PC recovery to it.
• The Q1 guide: Zinsner said a similar revenue hit is expected in the first quarter (biggo.com).

Hot take: the Q1 guide, not Q4, decides whether this was a bottom. If Intel holds revenue and margin there, the AI server story reasserts itself. If not, expect another ugly day.