Intel shares jumped after the company delivered a forecast that left analysts scrambling to revise their models upward. The numbers told a story of data-center demand outstripping supply. Customers building out AI infrastructure simply needed more chips than Intel could ship fast enough.
That scarcity translated into higher average selling prices. Revenue guidance for the third quarter topped Wall Street expectations. The stock reacted immediately. Yet the optimism came mixed with fresh concerns over capital spending.
Forecasts that outran estimates
Intel now expects to spend more than $20 billion on capital expenditures this year. The previous plan called for spending to stay roughly flat. Management pointed to long-term agreements and visible demand from AI workloads as justification. Yahoo Finance captured the immediate market reaction: shares rose on the news, signaling investors saw payoff from years of restructuring.
But not everyone cheered the higher outlay. Some analysts questioned whether the spending would deliver returns quickly enough. Foundry losses narrowed yet remained deep at $2.09 billion in the latest quarter. Progress on Intel 18A process technology showed yields improving and costs dropping. Panther Lake manufacturing costs had fallen about 50 percent so far this year. Another 20 percent reduction looked likely by year-end. Still, external foundry revenue stayed modest.
The Reuters report on July 23 offered more color. Intel forecast quarterly profit and revenue above estimates. Shares climbed 5.2 percent in after-hours trading. CEO Lip-Bu Tan has pushed the company toward rack-scale AI systems and next-generation Xeon processors. Reuters quoted Shay Boloor of Futurum Group: “The stock can continue revaluing if Intel converts the current data center shortage into sustained revenue growth, improves foundry economics and finally announces the external customers needed to validate the next stage of the manufacturing turnaround.”
Analysts piled on with target increases. At least six firms lifted price targets after the results. The median target sat 8.8 percent above the recent close, per LSEG data. Wells Fargo, JPMorgan, Rosenblatt and others saw upside tied to AI CPU momentum. Wedbush raised its target to $98 from $60. Rosenblatt moved to $80 from $65. The consensus rating across 31 analysts remains a hold, with an average price target near $100.61 according to Public.com.
Some forecasts proved more bullish. LongForecast projected Intel could reach $139 by the end of July before pulling back. Other models pointed to a 2026 range between $85 and $115. Those numbers assumed successful execution on foundry wins and sustained data-center growth. Reality has proven more volatile.
Earlier this month semiconductor stocks sold off broadly. Fears over AI spending slowed the rally. Intel fell alongside peers yet trades above some price targets. Forbes noted the stock sat 8 percent above its average target even after a 21 percent drop in early July. Morgan Stanley called the decline a mid-cycle reset rather than the end of the boom.
And the AI story keeps evolving. Intel unveiled new innovations at Computex this year. Xeon 6+ processors built on Intel 18A target high-density workloads. Core Ultra Series 3 processors power more than 325 PC designs. The platform now extends into handheld gaming with Arc G-series chips. Intel Newsroom highlighted strong customer uptake and improved 18A yields accelerating the roadmap.
Gaudi 3 accelerators gained traction for inference. The company pivoted from training ambitions after earlier generations fell short. CRN reported in May that Intel’s AI chip strategy under Tan now centers on GPUs, inference, and agentic computing. Rack-scale solutions replaced a canceled Falcon Shores GPU. Anil Nanduri, vice president of product management for data center AI accelerators, admitted past shortcomings at CES. “We did not meet the needs of the frontier AI training market, and we didn’t meet the market needs.”
Recent X discussions captured the tension. One analyst thread noted Q2 server revenue grew on 48 percent higher ASPs while shipment volumes rose only 9 percent. Customers paid up for scarce high-end Xeons. Long-term supply agreements brought in $1.7 billion in customer deposits. That cash signaled commitment. Demand existed. Supply constraints persisted.
But foundry economics still pressured margins. External revenue came in at $293 million, partly from reclassified Altera sales. 14A process node remains on the roadmap, yet big external customers have not yet materialized publicly. Investors weighed these risks. The stock rose sharply after earnings, then gave back gains the next session. One post on X described the reaction as applause for confirmed AI CPU demand followed by repricing of heavy capex and ongoing losses.
Intel’s turnaround hinges on three pillars. First, convert data-center CPU shortages into consistent revenue growth. Second, prove foundry viability beyond internal use. Third, secure external customers at scale. Government support helps. The CHIPS Act and state incentives back domestic manufacturing goals. Tan has strengthened the balance sheet and courted major investors.
D.A. Davidson analysts saw the capex hike as evidence of customer acquisition. “The aggressive capex raise is a proof point that Intel is likely to see continued customer acquisition as the United States demands more domestic semiconductor manufacturing.” Melius Research agreed. The spending increase reflected confidence in both product demand and foundry traction for packaging and 14A wafers.
Challenges remain. Nvidia still dominates accelerators. AMD offers competitive server CPUs. Intel must execute flawlessly on process technology. 18A yields have improved. Costs trend down. Yet the curve must bend further and faster to achieve attractive economics. Panther Lake progress offers hope. So does the shift toward open systems and rack-scale offerings.
Market sentiment swings with each data point. One day the narrative celebrates AI tailwinds. The next it obsesses over cash burn. Shares have more than doubled this year from earlier lows. They sit well off recent record highs. Volatility reflects the high stakes.
Longer term, analysts see upside if milestones are met. A $150 target appeared in some notes earlier this summer, contingent on major foundry wins and margin gains. Base cases cluster lower, around $85 to $110. The spread shows how much execution risk remains baked into the valuation.
Intel no longer competes only on raw transistor counts. It sells systems. It sells software stacks. It sells manufacturing capacity to others. That broader portfolio could cushion cyclical swings. It could also complicate investor focus. The market rewards clarity. Intel’s story still carries footnotes.
Recent coverage from CRN traced the evolution. After years of missteps with discrete AI accelerators, the company now emphasizes inference, agentic workloads, and integrated rack solutions. Cancellation of Falcon Shores marked a strategic pivot toward system-level competition. Gaudi 3 targets cost-effective inference against Nvidia’s H100 with claimed gains in throughput and price-performance for models such as Llama 2 70B.
Executives speak of AI everywhere. From PCs to edge to data centers. Lunar Lake processors pack more than 100 TOPS. Core Ultra chips drive PC refresh cycles with AI features. Handheld gaming devices add another vector. The bet is that AI proliferates across devices and workloads. Intel aims to supply pieces of that expansion.
Whether those pieces generate enough profit to offset foundry investment is the open question. Q2 results showed server strength. Foundry losses narrowed on one-time items more than operational improvement. Deposits from customers provide a buffer. Sustained volume growth must follow.
Investors will watch the October earnings closely. Guidance then could reset expectations again. New customer announcements would move the needle. Further details on 18A commercialization and external design wins would build credibility. Until then the stock trades on hope, data points, and macro sentiment toward AI capex.
The semiconductor cycle has always been brutal. This AI chapter feels different. Demand signals look structural. Supply responses require billions and years. Intel sits in the middle of that tension. Its recent forecast reminded the market that the company still matters in the AI buildout. The real test will be whether that momentum compounds or fades under the weight of its own ambitions.
WebProNews is an iEntry Publication