On July 23, 2026, after the US stock market closed (Beijing time), Intel (INTC) released its financial results for the second quarter of fiscal year 2026, ending June 27. The significance of this report was clear the moment the data was published—revenue reached $16.1 billion, up 25% year-over-year, marking the fastest single-quarter growth since Q3 2011. Adjusted earnings per share came in at $0.42, double the market expectation of $0.21. Non-GAAP gross margin was 41.8%, up a substantial 12.1 percentage points from the same period last year.
Following the release, Intel’s stock price surged more than 13% in after-hours trading. However, during the regular session on July 24 (Beijing time), Intel closed at $100.23, down 2.33% for the day. This swing highlights that the market’s pricing of this "beat expectations" report is not unilaterally optimistic—significant differences remain.
The real question is: What does this earnings report actually signify? Over the past two years, the spotlight of the AI boom has focused almost entirely on Nvidia, AMD, and Broadcom. GPUs have been seen as the sole narrative for AI computing power. Yet Intel’s data center business posted $6.3 billion in revenue, up 59% year-over-year, rewriting that story. Demand for AI infrastructure is now spreading from GPUs to a broader chip supply chain.
Data Center: From Supporting Role to Growth Engine
In Q2, Intel’s Data Center and AI (DCAI) division generated $6.3 billion in revenue, soaring 59% year-over-year and 24% quarter-over-quarter, setting a record for the strongest quarterly growth in server business history. This growth rate is more than double Intel’s overall revenue growth and easily beats analyst expectations of $5.6 billion.
Intel CEO Pat Gelsinger stated directly on the earnings call, "In the data center, CPUs are taking off." He further noted that demand is outpacing Intel’s increasing supply capacity. CFO David Zinsner added that pricing conditions are better than the company anticipated—in the Chinese market, some server CPU products have seen cumulative price increases of over 40% since the start of 2026.
The structural force driving this growth lies in the shifting center of gravity within the AI industry. Over the past two years, the market’s focus has been on high-performance GPU accelerators needed for AI model training. But as enterprises move from training large models to deploying "Agentic AI" at scale—AI agents capable of autonomous tasks—demand for inference-side computing power is exploding. At this stage, AI servers no longer rely solely on GPUs—CPUs, as the core scheduling hub, are seeing a synchronized surge in demand.
TrendForce research points out that AI applications are shifting from training to inference and Agentic AI, elevating server CPUs from auxiliary devices to central scheduling hubs. Intel’s Xeon 6 series processors have become one of the fastest ramping products in company history. Meanwhile, AMD’s EPYC server processors are also benefiting from this trend, with analysts projecting AMD’s 2026 server CPU shipments will grow by 15% to 20%.
In other words, the AI data center is a complex system composed of GPUs, CPUs, network chips, and storage. As demand for computing power spreads from training to inference, every link in the supply chain stands to benefit.
Is Intel Regaining Market Confidence?
Beyond the numbers, the market is more concerned about the sustainability of Intel’s current recovery.
From a fundamentals perspective, several positive signals emerge. First, Intel has beaten its financial guidance for seven consecutive quarters and has posted Non-GAAP profits for four straight quarters. Second, the company has signed 10 long-term supply agreements with server CPU customers—some locking in prices, others securing purchase volumes, with several contracts spanning three to five years. This long-term model provides greater visibility for future revenue.
Third—and most importantly—Intel is regaining pricing power. Supply shortages are evident not only in sales volumes but also in prices. Zinsner made it clear that "pricing is better than expected," and the more than 40% cumulative price increase for some CPU products since the beginning of the year is direct evidence of this.
However, concerns remain. The foundry business (Intel Foundry) reported Q2 revenue of $5.8 billion, up 31% year-over-year, but still posted an operating loss of $2.1 billion. More importantly, current foundry orders are mainly from Intel’s own product divisions, not external customers. While cybersecurity firm Fortinet became the first publicly disclosed foundry client since Gelsinger took over, the partnership uses mature manufacturing processes rather than Intel’s most advanced 14A node.
On advanced nodes, Intel’s 18A process has reached mass production with yields above expectations, and development of the next-generation 14A node is on track. The company is "fully committed" to large-scale production using the 14A process in 2028. Yet whether the foundry business can attract enough major external customers remains the key variable for Intel’s long-term transformation.
Semiconductor Cycle: From Compute Shortage to Infrastructure Expansion
Zooming out from Intel to the broader semiconductor industry, a bigger question arises: Is the semiconductor cycle entering a recovery phase?
From 2024 to 2025, AI compute shortages dominated the industry—GPU supply lagged, HBM memory prices soared, and advanced packaging capacity was stretched thin. Entering 2026, the narrative is shifting. Leading global cloud providers and AI companies continue to raise their infrastructure investment budgets. Nomura forecasts global AI server revenue will rise 78% in 2026 and 76% in 2027. WSTS expects the global semiconductor market to expand by more than 25% in 2026.
Intel’s management offered a more specific outlook on the earnings call: server CPU industry shipments are expected to maintain strong double-digit growth in 2026 and 2027, with the boom cycle extending through 2028. Gelsinger emphasized that the core industry challenge is "demand far outstripping supply," with persistent shortages in wafers, substrates, memory, and advanced packaging that cannot be resolved in the short term.
Based on this assessment, Intel announced it would raise its 2026 capital expenditure forecast to over $20 billion and signaled that 2027 capex will significantly exceed 2026 levels. The company revealed that from 2021 to 2026, its combined capital spending on tools and facilities in the US approached $100 billion.
From a cyclical standpoint, the semiconductor industry is currently in a phase where "inventory clearing and capacity expansion coexist." Surging demand for AI inference is driving both volume and price increases in chip design. The industry is transitioning from "unit price increases" to "capacity expansion."
Of course, an upward cycle doesn’t mean there are no risks. Morgan Stanley’s latest report warns that the AI-driven boom in semiconductor memory is nearing an inflection point, with contract memory prices expected to peak in Q4 2026. The price trajectory of memory chips is often seen as a leading indicator for the semiconductor cycle—if memory prices peak first, does that signal a broader chip cycle turning point? This is a variable worth monitoring.
Conclusion
Intel’s Q2 2026 earnings report is not just a case of one company beating expectations—it marks a pivotal moment as demand for AI computing power spreads from GPUs to a broader chip supply chain. The "takeoff" of data center CPUs, the return of pricing power, and the sharp increase in capital expenditures all point to one conclusion: the AI chip cycle is moving from a phase of "compute shortages" to a new stage of "infrastructure expansion."
Naturally, the evolution of the cycle is never linear. The expansion of foundry business to external clients, the potential inflection point in memory prices, and the sustainability of tech giants’ capital spending are all uncertainties that require ongoing attention. But for Intel and the entire semiconductor value chain, the spread of AI demand from GPUs to CPUs, network chips, and storage is opening up a much broader growth opportunity than the narrow "GPU narrative" alone.
FAQ
Q1: What is the most critical "beat expectations" data in Intel’s Q2 earnings?
Intel’s Q2 revenue reached $16.1 billion, up 25% year-over-year—the fastest growth since Q3 2011 and far above the market expectation of $14.4 billion. Adjusted EPS was $0.42, double the expected $0.21. Data center business revenue was $6.3 billion, up 59% year-over-year. The company has beaten its financial guidance for seven consecutive quarters.
Q2: Why is Intel’s CPU business suddenly growing so fast?
The AI industry is shifting from model training to large-scale deployment of "Agentic AI." Exploding demand for inference-side computing power is turning general-purpose CPUs from auxiliary devices into the central scheduling hub for AI servers. AI data centers now require not only GPUs but also large numbers of CPUs to support servers, storage, networking, and inference workloads. Intel’s Xeon 6 series has become one of the fastest ramping products in company history.
Q3: Can Intel’s current recovery be sustained?
The company expects server CPU industry shipments to maintain strong double-digit growth in 2026 and 2027, with the cycle continuing through 2028. Intel has raised its 2026 capital expenditure to over $20 billion and signaled further increases in 2027. However, whether the foundry business can attract enough major external customers remains a key variable. The risk of memory prices peaking in Q4 2026 also needs to be monitored.
Q4: What stage is the AI chip cycle currently in?
The industry is moving from the "AI compute shortage" phase of 2024–2025 into a new stage of "AI infrastructure expansion." Leading global cloud providers are continually raising their infrastructure investment budgets, and Nomura projects global AI server revenue will grow 78% in 2026. The semiconductor industry is now in a phase where inventory clearing and capacity expansion coexist, with surging AI inference demand driving both volume and price increases in chip design.
