The sales growth was primarily driven by Intel's data center and artificial intelligence (AI) division, with revenue from this division increasing by 22% to reach $5.1 billion.
"In the past few years, the topic in the field of high-performance computing has been almost entirely focused on graphics processors and other accelerators," Lip-Bu Tan, the chief executive, said on an earnings call on Thursday afternoon. "But in recent months, we have seen clear signs that CPUs are becoming indispensable again as the foundation for the AI era.". "
As more and more AI computing workloads shift from training to inference (i.e., deploying AI models to perform tasks), CPUs have gained greater application.
"I believe that in terms of orchestration, controlling the plane, and managing data from all different agents, CPUs are much more efficient," Chen said.
"I believe that the CPU business will continue to be an important growth engine for the company in the coming years, not just in the next few quarters," he added.
Intel's largest revenue source, the client computing group business (including sales of personal computer chips), saw a 1% increase in revenue to $7.7 billion.
Intel previously stated that due to continuously growing demand, the company faces chip supply constraints, and therefore prioritizes serving customers of data centers and artificial intelligence products over those of personal computer products.
"Even though we have increased our factory output, the demand for all our business segments still outstrips supply, especially for Xeon server CPUs. We anticipate that this segment will maintain its growth momentum this year," said Mr. Chen.
Intel's Chief Financial Officer David Zinsner said supply constraints may begin to ease as the company "supplied at the lowest point of the year in the first quarter".
According to a previous report by Nikkei Asia, the supply shortage of CPUs produced by Intel and AMD is worsening, adding to the already unprecedented impact of memory chip shortages on PC and server manufacturers. The supply shortage has led to longer waiting times for CPU deliveries for servers and PCs, as well as price increases.
Intel had previously forecasted that its revenue for the April to June period would reach between $13.8 billion and $14.8 billion.
Meanwhile, Intel's chip foundry business, despite experiencing revenue growth, continued to drag down the company's profits. According to Intel's wafer foundry business, revenue for the three months from January to March increased by 16% year-on-year to US$5.4 billion. However, due to the high cost of expanding wafer fabs, operating profits suffered a loss of US$2.4 billion.
The company said it had expanded its assembly and testing capacity in Penang, Malaysia, during the quarter.
"This expansion will help meet identified demand, which will begin to translate into revenue in 2027," said David Zinsner, Intel's chief financial officer, on Thursday
Intel Foundry has also secured a new customer: Elon Musk.
The tech billionaire said Intel would be the manufacturing partner for his Terafab project, a joint venture between Tesla, SpaceX and xAI, which aims to produce 1m wafers a month.
Musk stated that this massive wafer fab will utilize Intel's 14A chip manufacturing technology, which is still under development.
"Both Elon and I firmly believe that the global semiconductor supply cannot keep up with the rapid growth of demand," Chen said on Thursday, adding that both parties are exploring "unconventional methods" to enhance chip manufacturing efficiency.
By the end of 2025, Intel will begin mass production of its 18A process at its Chandler, Arizona, factory - its most advanced process in mass production - with the goal of competing with TSMC's 2-nanometer technology, the world's top wafer foundry.
Intel said on Thursday that the "progress of 18A and the next-generation 14A manufacturing technology has exceeded expectations".
"A year ago, people were discussing whether Intel could survive. Today, the focus of discussion has shifted to how we can increase production capacity and expand supply scale as soon as possible to meet the huge demand for our products in the market," Chen said. "Intel today is a completely different company compared to what it was a year ago."
Intel has been a Wall Street darling of late, with its shares up more than 80 per cent this year through Thursday's close, after surging 84 per cent in anticipation of 2025. The chipmaker has been courted by the Trump administration, which last year became Intel's largest shareholder in a move aimed at bringing chipmaking back to the US. Nvidia and SoftBank have also invested billions of dollars in Intel.
However, this company lagged far behind its competitors NVIDIA and AMD in the early days of the AI boom, and has yet to show much momentum.
This situation may finally be changing. Revenue increased by 7.2% compared to the same period last year, which was $12.67 billion. Previously, revenue had declined year-on-year in five of the past seven quarters.
Intel expects second-quarter revenue to be between $13.8 billion and $14.8 billion, with adjusted earnings per share of $0.20. This is significantly higher than analysts' previous expectations of $13.07 billion in revenue and $0.90 in earnings per share.
Intel's data center business experienced the strongest growth, benefiting from surging demand for central processing units (CPUs) as the company began to make progress in the field of artificial intelligence. Revenue from this segment increased by 22% to reach $5.1 billion.
The once-dormant CPU market is now thriving, as the emergence of intelligent workloads has altered computing demands, no longer confined to the Nvidia Graphics Processing Unit (GPU) that previously dominated the AI sector. This growing demand for CPUs prompted Intel to recently acquire a 49% stake in its Irish chip manufacturing facility for $14 billion, which had previously been sold to Apollo Global Management.
"CPUs are becoming indispensable again as the foundation of the AI era," said Intel CEO Brian Krzanich on the company's earnings call. "This is not just our wishful thinking, but feedback we have heard from our customers.". "
Intel has adopted an unusual strategy in the chip field. As a comprehensive equipment manufacturer, Intel not only produces its own products but also produces the chips that drive these products. Most chip manufacturers outsource the complex and costly manufacturing processes to large chip manufacturers such as TSMC.
Intel's foundry revenue grew 16 per cent year-on-year to $5.4bn, even though the bulk of its foundry business is making its own chips.
The Intel Core Ultra series 3 processor began selling in the PC market in January, while its latest Xeon 6+ data center processor was launched in March. Shortly thereafter, Google pledged to use multiple generations of Intel CPUs in its data centers to run AI workloads.
Intel's latest PC and data center processors are manufactured using the 18A process node at a newly built giant wafer fab in Arizona. Currently, despite the technical similarity between the 18A process node and TSMC's 2nm process node, Intel remains the sole major customer of its 18A chip wafer fab.
The real challenge lies in persuading TSMC's long-standing customers to make this transition.
Intel is recovering from years of delays in several previous process nodes, where some 18A wafers had defects, resulting in a reduction in the number of usable chips per wafer, commonly referred to as yield.
Some analysts are waiting for Intel's next-generation 14A technology to demonstrate satisfactory yield rates, with plans to launch it in 2028 or later. Previously, Intel CEO Brian Krzanich had stated that Intel would wait for major customers to emerge before investing heavily in upgrading to the latest technology. However, in January this year, he said at the X Forum that Intel was "aggressively advancing 14A technology".
During the earnings call, Chen Liwu stated that "numerous customers" are "actively evaluating this technology," and that its development is progressing at a faster pace than Intel's 18A technology.
One important customer may be Elon Musk, but the specific details remain unclear. Intel announced earlier this month that it will join Musk's Terafab chip manufacturing facility in Austin, Texas, to help SpaceX, xAI, and Tesla "design, manufacture, and package ultra-high-performance chips on a large scale.".
During Tesla's first-quarter earnings call on Wednesday, Musk stated that Tesla plans to use Intel's upcoming 14A process to produce chips at the factory, which will manufacture chips for Tesla cars and robots, as well as for the orbital data center that SpaceX has yet to build.
Musk stated that Intel is still developing 14A, but "by the time Terafab scales up, 14A may already be quite mature or ready for use.".
"Both Elon and I are convinced that global semiconductor supply can't keep up with the rapid growth in demand," Mr Chen said on Intel's conference call, adding that they were working together to "find unconventional ways to improve manufacturing efficiency".
Intel's renewed focus on manufacturing chips for other companies began after Pat Gelsinger took over as CEO in 2021. Gelsinger stepped down in 2024 and was succeeded by Tan early last year.
The latest guidance is so powerful possibly because Intel excels in another advantageous area of chip manufacturing processes: advanced packaging, which involves connecting individual chips into a larger system. Intel is one of only three companies globally that offer the most advanced packaging technology, creating a new bottleneck in the competition for AI chips.
In an interview with CNBC, Chief Financial Officer David Zinsner stated that he believes advanced packaging technology will bring billions of dollars in revenue to each customer, whereas his previous estimate was only in the hundreds of millions. Intel's advanced packaging customers include Amazon, Cisco, as well as new clients SpaceX and Tesla.