Intel reported its fastest revenue growth in 15 years and added $2 billion (€1.75 billion) to its capital spending plans yesterday as demand for chips used in AI data centres accelerated its financial turnaround.
Revenue rose to $16.1 billion in the three months to the end of June, up 25 per cent year on year, and Intel forecast $15.8 billion to $16.8 billion in sales for the current quarter, far exceeding Wall Street expectations.
The US group has rebounded from several difficult years after receiving backing from the Trump administration for its costly push to regain its status as America’s only world-class processor chipmaker.
Intel on Thursday also upped its planned capital spending for this year from $18 billion to $20 billion and forecast higher spending into 2027. The stock rallied 4 per cent in after-hours trading.
RM Block
Intel’s shares raced to all-time highs this spring after bumper earnings three months ago but have retreated about 25 per cent in the past month alongside other chip stocks as investors worry about the sustainability of the AI rally.
“AI is driving unprecedented demand for compute, and as we continue to execute, Intel is well positioned to capture sustainable growth,” said Intel chief executive Lip-Bu Tan. The “new Intel” was taking shape, he added.
The company’s revenue growth for the June quarter was largely driven by demand and price adjustments for Intel-designed chips, with its product revenue hitting $15.1 billion, well above estimates of $13.6 billion.
Foundry revenue, relating to its critical manufacturing business, beat expectations at $5.8 billion, up 31 per cent from a year ago.
Intel’s data centre and AI business revenue – relating to the server chips used in the infrastructure that trains and runs AI models – was up 59 per cent to $6.3 billion, above the $5.6 billion analysts expected, according to Visible Alpha.
Having faced an uncertain future only a year ago, Intel is gaining ground thanks to surging demand for its central processing units and the promise of new clients for its semiconductor manufacturing business.
Intel, alongside chip designers AMD and Arm, has pointed to a greater demand for CPUs in the infrastructure needed to run AI applications through a process known as “inference”.
CPUs have grown in importance relative to the graphics processing units (GPUs) specialised in the large parallel computations needed to train advanced AI models, which dominated the first stage of the AI infrastructure boom.
AMD chief executive Lisa Su on Thursday raised the company’s projections for how large the CPU market will be by the end of the decade to $220 billion, from a recent $120 billion estimate.
The Trump administration has meanwhile thrown its weight behind Intel as the US’s only contender to match Taiwan Semiconductor Manufacturing Company as a maker of top-flight processor chips. The US government struck a deal to take a 10 per cent stake in the company last year.
Tan has since secured multibillion-dollar investments from Nvidia and SoftBank and announced plans to work with Elon Musk on his “Terafab” manufacturing project, all while pursuing companywide cuts.
Adjusted net income rose to $2.2 billion. Net income under the generally accepted accounting principles swung to an $11 billion loss because of volatility around shares held in escrow under Intel’s investment deal with the government.
Intel aims to compete with Nvidia and AMD with its own AI accelerator chip later this year and is trying to lure big manufacturing clients such as Apple away from TSMC.
Only last year, Tan had warned Intel might abandon its latest manufacturing process – known as 14A – if it were unable to land customers to justify the large investment.
Trump in June pre-emptively announced that Apple would start using Intel as a manufacturer in June, a claim that neither company has confirmed.
Tan on Thursday said the company was “fully committed” to high-volume production on 14A in 2028. – Copyright The Financial Times Limited 2026
