By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is bearish on Intel, with concerns about the company's foundry losses, high capex requirements, and the cyclical nature of its CPU pricing power. The panelists also note that Intel's AI leadership is not yet sustainable, and the stock's high forward P/E ratio may not reflect a margin of safety.
Risk: The panelists flagged the potential for Intel's foundry segment to remain a drag on earnings and the risk of elevated capex requirements without significant market share capture in the foundry business.
Opportunity: The potential opportunity lies in the possibility of U.S. government subsidies (CHIPS Act) absorbing losses and guaranteeing volume, which could shift the breakeven timeline for Intel's 18A node.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Better late than never: Intel (NASDAQ: INTC) has officially joined the artificial intelligence (AI) infrastructure boom, with the company seeing its data center and AI segment revenue surge in the second quarter. Shares jumped in after-hours trading, but Intel gave back those gains after management acknowledged the company would meaningfully increase its capital expenditures (capex) this year. Punishing AI stocks that increase their capex has been a major theme this earnings season.
Intel's stock has now lost about a third of its value from its recent highs. However, its shares are still up more than 150% on the year and over 300% in the past 12 months.
Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »
The biggest driver of Intel's growth was data center central processing units (CPUs). Demand for server CPUs has increased in the second quarter on a sequential basis and management expects double-digit industry growth both this year and next, with momentum continuing into 2028. Meanwhile, demand continues to far outstrip supply, leading to higher CPU prices across the board.
This helped lead to a 59% jump in its data center and AI product revenue to $6.3 billion. Its client computing group product revenue rose 13% year over year to $8.9 billion, leading to total product revenue climbing 28% to $15.1 billion. That was a big step up from the 9% rise in product revenue it saw in Q1.
Intel's foundry business, meanwhile, saw revenue surge 31% to $5.8 billion. However, the segment continues to post large operating losses, with a $2.1 billion loss in the quarter. Revenue from Intel's other businesses sank 33% year over year to $0.7 billion, largely due to the sale of 51% of its Altera subsidiary.
Overall revenue for Intel climbed 25% to $16.1 billion, its fastest growth in almost 15 years. Its adjusted earnings per share went from a loss of $0.10 to a profit of $0.42. This was bolstered by a big 1,290 basis-point increase in its gross margins to 40.4%.
Looking ahead, Intel projected Q3 revenue to be between $15.8 billion and $16.8 billion with adjusted EPS of $0.38. That was well ahead of the $0.27 in EPS and $15.1 billion in sales that analysts were expecting. It projected gross margins to be around 41%.
It expects supply constraints to affect it next quarter, with improvements starting to show up toward the end of the quarter and into Q4. It also raised its projected capex, taking its 2026 budget from initial guidance of around $17 billion to $18 billion to over $20 billion. Capital expenditures for 2027, meanwhile, are expected to be significantly above 2026 levels.
While Intel is starting to see strong momentum in its data center and AI segment, this seems more driven by the sudden need of hyperscalers to grab whatever server CPUs they can to help handle agentic AI than by any big moves that the company has made. CPU demand is through the roof, which is helping with pricing in its client computing group segments. And while its foundry business is gaining some traction, it remains a money-draining business during one of the biggest semiconductor booms of all time.
Before its run, Intel was a cheap stock barely trading above the value of its physical assets. Now it has a forward P/E of 105, and the stock's valuation looks bloated for a company that seems more like a passenger than one helping to drive the market. I think there are better AI stocks out there to buy.
Ever feel like you missed the boat in buying the most successful stocks? Then you’ll want to hear this.
On rare occasions, our expert team of analysts issues a “Double Down” stock recommendation for companies that they think are about to pop. If you’re worried you’ve already missed your chance to invest, now is the best time to buy before it’s too late. And the numbers speak for themselves:
Right now, we’re issuing “Double Down” alerts for three incredible companies, available when you join Stock Advisor, and there may not be another chance like this anytime soon.
**Stock Advisor returns as of July 27, 2026. *
Geoffrey Seiler has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Intel. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Even with genuine AI tailwinds, Intel's 105x forward P/E leaves no margin of safety for execution slips or capex creep."
Intel's Q2 beat was genuine: data-center/AI revenue +59% to $6.3B, total revenue +25% to $16.1B, gross margin expansion of 1,290bp to 40.4%, and EPS swing from -$0.10 to +$0.42. Guidance also topped consensus. Yet the market sold the news because 2026 capex jumped from ~$17B to >$20B and 2027 will be "significantly" higher. At 105x forward P/E, the stock prices in flawless execution on foundry losses, Gaudi accelerators, and sustained CPU pricing power. The article correctly flags Intel as riding hyperscaler desperation rather than leading the AI wave; history shows late-cycle capex spikes often destroy returns when supply catches up.
If Intel's foundry finally hits breakeven by 2027 and its 18A process node proves competitive, the re-rating from 105x to a more normal 25-30x on 15-20% EPS growth could still deliver 2x+ upside over 2-3 years; the article underweights that optionality.
"Intel's current valuation is disconnected from its reality as a legacy CPU vendor struggling to fund a capital-intensive foundry turnaround."
The article’s excitement over a 59% revenue jump in data center products misses the structural decay in Intel's foundry business. A $2.1 billion operating loss in the foundry segment during a global semiconductor supercycle is a red flag, not a growth story. While the CPU pricing power is real, it is a cyclical tailwind, not a fundamental pivot to AI leadership. With a forward P/E of 105, the market is pricing Intel as a high-growth AI powerhouse, but the reality is a legacy firm struggling with massive, escalating capex requirements of $20B+ annually. You are paying a premium for a company that is essentially subsidizing its own manufacturing turnaround.
If Intel's foundry strategy succeeds in attracting high-margin external customers, the current massive capex will look like a bargain, transforming the company into a critical geopolitical asset with an unassailable moat.
"Intel is a beneficiary of a supply shortage, not an AI infrastructure leader, and its 105x P/E assumes foundry profitability that remains unproven and years away."
Intel's 59% data center revenue surge is real, but the article conflates two separate stories: (1) cyclical CPU supply shortage driving price increases, and (2) sustainable AI leadership. The foundry segment losing $2.1B while generating $5.8B revenue is the canary—Intel is burning cash to compete in a capital-intensive race where TSMC and Samsung already lead. Capex rising to $20B+ in 2026 and 'significantly above' that in 2027 signals management expects years of losses ahead. A 105x forward P/E on cyclical upside is not a margin of safety.
If hyperscalers are genuinely supply-constrained through 2027-2028 and Intel's foundry capex actually closes the process node gap with TSMC by 2026-2027, Intel could capture sustained market share at premium pricing—turning foundry from a loss center into a profit engine within 18-24 months.
"Intel's upside hinges on a durable AI-driven demand and a cash-flow-positive foundry; otherwise heavy capex and a sky-high valuation leave substantial downside risk."
Intel's quarterly ascent looks impressive on paper: data center AI revenue up 59% to 6.3B, gross margins expanding to 40.4%, and a higher capex plan signaling ambition to gain share. But the headline numbers mask fragility: much of the AI uplift appears tied to hyperscalers scrambling for CPUs amid supply tightness, not a durable demand engine. The 41% gross-margin target in 2024-26 seems optimistic if capex remains elevated and the foundry segment—still loss-making—persists as a drag. The stock trades at a forward P/E around 105, and the path to meaningful earnings leverage hinges on sustained AI infra demand and a cash-flow-positive foundry. Risks loom if demand cools or if transitions delay.
If AI compute demand proves durable and Intel wins meaningful foundry share, the stock could re-rate despite today’s high multiple. The article’s caution on capex ignores potential long-run margin improvement from foundry customers and possible subsidies or multi-year contracts.
"Foundry losses are investment cost, not permanent decay, if 18A lands external customers."
Claude and Gemini both treat the $2.1B foundry loss as structural decay, yet ignore that Intel's 18A node is the first credible external-foundry offering in years. The capex spike is the price of catching up; if it succeeds, the current 105x P/E embeds the risk premium that evaporates on first major external win. Nobody has quantified what even one $3-4B annual foundry contract does to the P&L trajectory.
"Intel's foundry strategy assumes a level of execution and market share capture from TSMC that lacks historical evidence, making the $20B+ capex a massive risk rather than a guaranteed moat."
Grok, you are betting on a 'first major external win' that has yet to materialize in any meaningful volume. The $2.1B foundry loss isn't just 'catching up' costs; it is a structural burn rate that assumes a market share capture from TSMC that has zero historical precedent for Intel. If Intel's 18A node doesn't yield significantly better than their recent history, that $20B+ capex becomes a sunk cost, not an investment.
"Government subsidies could justify capex that private-market ROI cannot, but the article and panel ignore this geopolitical wildcard entirely."
Grok's '$3-4B foundry contract' math is illustrative but assumes Intel wins at TSMC-competitive yields and pricing simultaneously—historically Intel's weakness. Gemini correctly flags zero precedent, but both miss the geopolitical angle: U.S. government subsidies (CHIPS Act) could absorb losses and guarantee volume, making traditional P&L math irrelevant. That changes capex risk assessment entirely, yet nobody quantified potential subsidy impact on breakeven timeline.
"Explicit subsidy scenarios are needed because CHIPS Act subsidies could materially alter capex economics, but timing and terms may blunt upside."
Claude raised subsidies can flip the math, but the piece doesn't quantify. If CHIPS Act capex subsidies or volume guarantees materialize, Intel's breakeven on 18A could shift earlier—but policy timing risk and potential clawbacks could delay payoffs. Even partial subsidies leave a high cost of capital for $20B+ annual capex, contingent on durable demand. The article should include explicit subsidy scenarios to test upside versus capex risk.
The panel consensus is bearish on Intel, with concerns about the company's foundry losses, high capex requirements, and the cyclical nature of its CPU pricing power. The panelists also note that Intel's AI leadership is not yet sustainable, and the stock's high forward P/E ratio may not reflect a margin of safety.
The potential opportunity lies in the possibility of U.S. government subsidies (CHIPS Act) absorbing losses and guaranteeing volume, which could shift the breakeven timeline for Intel's 18A node.
The panelists flagged the potential for Intel's foundry segment to remain a drag on earnings and the risk of elevated capex requirements without significant market share capture in the foundry business.