Outrageous Predictions
Révolution Verte en Suisse : un projet de CHF 30 milliards d’ici 2050
Katrin Wagner
Head of Investment Content Switzerland
Artificial intelligence demand is lifting Intel’s server-chip business faster than its personal-computer operations.
Better factory output and stronger margins suggest execution is improving, not merely demand.
The foundry remains the decisive test because most of its revenue still comes from manufacturing Intel’s own chips, rather than winning outside customers.
Intel’s comeback is no longer running only on hope. On 23 July 2026, the chipmaker reported a stronger quarter than expected and raised its outlook. Demand for data-centre processors surged, factories produced more usable chips and profitability improved.
The shares jumped after the announcement, then gave back much of the initial gain. Intel showed progress, but the market has already rewarded the comeback story heavily this year. A good quarter confirms the direction. It does not settle the destination.
Intel designs processors and owns factories that manufacture them. Its central processing units, or CPUs, are the general-purpose brains inside computers and servers.
Graphics processing units, or GPUs, supplied by companies such as Nvidia receive most of the attention. They handle the heavy mathematical work behind artificial intelligence models. Yet data centres also need CPUs to organise data, manage applications and keep systems working.
That role becomes more valuable as artificial intelligence moves from training models towards using them in everyday tasks. More applications and automated agents mean more general computing around each accelerator.
Intel’s total revenue rose 25%, while its data-centre and artificial intelligence division grew 59%, according to the company’s second-quarter results, as reported by Bloomberg. Management also said demand was running ahead of supply. For the broader industry, this suggests the boom is widening beyond one product category. The orchestra still has a famous soloist, but the rest of the band is getting louder.
Strong demand only helps when a company can deliver. Intel has spent years struggling with factory delays, weak output and expensive expansion plans.
Intel said manufacturing yields improved. Yield means the share of chips produced that work well enough to sell. Better yields reduce waste, increase supply and spread factory costs across more products.
Intel’s gross margin, the share of revenue left after production costs, rose to around 40%. That remains below its strongest years, but the direction matters. The company also moved from an operating loss to an operating profit.
The headline accounts showed a very large net loss, but most came from an accounting revaluation linked to shares held for the United States government. The underlying operations were profitable. Accounting can occasionally make a recovering patient look as if it has fallen down the stairs.
Intel wants to manufacture chips for outside customers, much as Taiwan Semiconductor Manufacturing Company does. This factory-for-hire business is called a foundry.
Foundry revenue grew strongly, helped by better output and internal demand. However, most sales still come from Intel’s own product divisions, and the unit remains deeply loss-making. Higher reported revenue is not yet the same as winning profitable external customers.
Intel said it has several customer discussions and is progressing with newer manufacturing technology. Investors now need harder evidence: named customers, firm production commitments and falling losses.
The company is also increasing investment. This may support growth, but factories consume cash long before they produce reliable profits. The question is not whether Intel can spend. History has answered that. It is whether each investment earns an acceptable return.
The first risk is that artificial intelligence spending slows before Intel earns enough from new capacity. Watch customer orders, supply conditions and management’s guidance.
The second is factory execution. Better yields must continue as Intel moves into more advanced technology. Delays, rising costs or weaker margins would challenge the recovery.
The third is expectation risk. Intel’s shares have risen sharply, so even strong results may disappoint when investors expect near-perfect progress. Business improvement and a good share price are related, but not identical twins.
Intel entered earnings needing to prove that its comeback was becoming a business result rather than a market story. It did that. Artificial intelligence demand is lifting server processors, factories are producing more effectively and margins are recovering.
The next stage is harder. Intel must improve supply, protect profitability and persuade outside customers to trust its factories. It must also fund that effort without allowing investment to outrun cash generation.
One strong quarter does not restore a fallen champion, but it changes the burden of proof. Intel no longer needs to show that recovery is possible. It now needs to show that recovery is repeatable. Comebacks attract attention. Endurance creates value.
This material is marketing content and should not be regarded as investment advice. Trading financial instruments carries risks and historic performance is not a guarantee of future results.
The instrument(s) referenced in this content may be issued by a partner, from whom Saxo receives promotional fees, payment or retrocessions. While Saxo may receive compensation from these partnerships, all content is created with the aim of providing clients with valuable information and options.