In the fast-paced world of tech, few companies have faced as much scrutiny and transformation as Intel. The recent Intel Q4 2025 concall provided a clear window into the chip giant’s progress amid ongoing challenges. Investors and industry watchers tuned in eagerly to hear from CEO Lip-Bu Tan about the company’s financial performance, strategic shifts, and future vision. This explores the key takeaways from the Intel Q4 2025 concall, focusing on the persistent foundry losses, the aggressive push into AI PCs, and Tan’s comprehensive turnaround plan that’s aiming to steer Intel back to dominance.
Setting the Stage: Intel’s Journey Leading Up to Q4 2025
Intel has long been a cornerstone of the semiconductor industry, powering everything from personal computers to data centers. But in recent years, the company has grappled with fierce competition from rivals like AMD, Nvidia, and TSMC. Manufacturing delays, market share erosion in key segments, and the explosive growth of AI have put immense pressure on Intel’s operations. By the end of 2024, Intel was reporting significant losses in its foundry business, which is the arm responsible for manufacturing chips not just for Intel’s own products but potentially for external customers as well.
Enter Lip-Bu Tan, who took the helm as CEO in March 2025 after a period of leadership transition. Tan, a veteran in the semiconductor space with a background in venture capital and executive roles at companies like Cadence Design Systems, was brought in to inject discipline and focus. His appointment came at a critical time, following the retirement of previous CEO Pat Gelsinger, whose ambitious “IDM 2.0” strategy—aiming to reclaim manufacturing leadership through rapid node advancements—had mixed results. Tan’s approach has been more pragmatic, emphasizing cost controls, ecosystem partnerships, and a laser focus on profitable growth.
The Intel Q4 2025 concall, held on January 22, 2026, was Tan’s latest opportunity to update stakeholders on the company’s fiscal year-end performance. With the stock having surged nearly 80% over the course of 2025 thanks to key investments from Nvidia and SoftBank, expectations were high. Tan didn’t disappoint, delivering a message of steady progress while acknowledging the road ahead remains tough.
Key Financial Highlights from the Intel Q4 2025 Concall
Let’s start with the numbers, as they set the tone for the entire discussion. Intel reported Q4 2025 revenue of $13.5 billion, beating the upper end of its guidance range of $12.8 billion to $13.8 billion. This marked a sequential increase from Q3’s $13.3 billion and a year-over-year growth of about 5%, driven primarily by stronger demand in the data center and AI segments.
Non-GAAP earnings per share came in at $0.10, surpassing the $0.08 guidance provided in the previous quarter. Gross margins improved to 38.5%, up from 36.5% in Q3, thanks to better product mix and cost efficiencies. However, the company still posted a GAAP net loss of $0.5 billion for the quarter, largely due to ongoing restructuring charges and foundry investments.
For the full year 2025, Intel’s revenue totaled $52.8 billion, a modest increase from 2024’s $51.2 billion. The Client Computing Group (CCG), which handles PC chips, contributed $28 billion, while the Data Center and AI (DCAI) segment brought in $18 billion. Intel Foundry Services (IFS), the manufacturing arm, generated $17.5 billion in revenue but continued to weigh on profitability with operating losses narrowing to $6.5 billion for the year—better than the $7.5 billion peak in 2024 but still a significant drag.
Tan highlighted these figures as evidence that the company’s execution is improving. “This marks our fifth consecutive quarter of beating guidance,” he said during the call. “We’re seeing the benefits of our disciplined approach, with stronger margins and cash flow generation.” Cash from operations was $3.2 billion in Q4, helping to reduce net debt and fund ongoing capital expenditures, which were capped at $25 billion for the year—a reduction from prior plans.
Analysts on the call pressed Tan on tariffs and supply chain disruptions, given the geopolitical tensions. He noted that customer pull-ins in anticipation of potential tariffs boosted Q4 sales, but emphasized Intel’s efforts to diversify manufacturing, including expansions in the U.S., Europe, and Asia.
Foundry Losses: The Elephant in the Room
One of the most discussed topics in the Intel Q4 2025 concall was the foundry business. Intel’s push to become a world-class foundry—manufacturing chips for others, not just itself—has been ambitious but costly. The segment, rebranded as Intel Foundry, has been hemorrhaging money as the company invests heavily in advanced process nodes like 18A (1.8nm) and the upcoming 14A (1.4nm).
In Q4, foundry revenue was $4.5 billion, up slightly from $4.4 billion in Q3, but operating losses stood at $1.8 billion. For the full year, losses totaled $6.5 billion, an improvement from 2024’s $7.5 billion peak. Tan attributed the narrowing losses to higher wafer volumes, better yields on mature nodes like Intel 7, and initial external customer wins.
“Why are these losses persisting?” an analyst asked during the Q&A. Tan responded candidly: “Building a foundry isn’t cheap or quick. We’re investing in leading-edge tech to compete with TSMC, but we’re seeing traction. Our 18A node is ramping ahead of schedule, with yields improving to over 60%, and we’ve secured commitments from major players.”
Indeed, 2025 saw Intel land key deals, including advanced packaging services for Nvidia and potential wafer production for other AI chipmakers. Tan reiterated the goal of achieving foundry break-even by midway through the decade, targeting 40% gross margins by 2030. To get there, Intel is flattening its organization, reducing headcount by 15% (bringing the workforce to around 75,000), and scrutinizing every expense.
Critics argue that the foundry model is a double-edged sword. On one hand, it leverages Intel’s massive $100 billion in capital assets for external revenue. On the other, it competes directly with TSMC, which dominates with over 50% market share. Tan countered this by emphasizing differentiation: “We’re not just a fab; we’re an end-to-end provider with advanced packaging like EMIB and Foveros, plus our x86 ecosystem.”
Looking ahead, Tan forecasted foundry revenue growth in 2026, driven by 18A production for internal products like Panther Lake AI PCs and external customers. He hinted at “big time” investments in 14A, but only if customer demand justifies it— a shift from past “blank check” spending.
The AI PC Push: Intel’s Bet on the Future of Computing
Shifting gears, the Intel Q4 2025 concall shone a spotlight on the company’s aggressive AI PC initiative. AI PCs, which integrate neural processing units (NPUs) for on-device AI tasks, represent Intel’s response to the AI boom. Tan described 2025 as a “defining year” for this category, with Intel shipping over 40 million AI PC processors by year-end, on track to hit 100 million cumulative units by the end of 2026.
The star of the show was the Core Ultra series, including the newly unveiled Panther Lake architecture—the first AI PC platform built on Intel’s 18A process. Panther Lake promises up to 3x AI performance gains over prior generations, with improved power efficiency for laptops and desktops. “We’re extending leadership in AI PCs,” Tan said. “From consumer devices to enterprise edge computing, our ecosystem now supports over 400 AI features.”
Q4 saw strong demand for Core Ultra 200V series in business laptops, with partners like HP launching models like the EliteBook X. These devices handle AI workloads locally, enhancing privacy and speed for tasks like video editing, real-time translation, and generative AI. Tan highlighted benchmarks showing up to 223% faster performance on AI apps compared to non-AI PCs.
But it’s not just hardware. Intel’s AI PC Acceleration Program, launched in late 2024, has partnered with over 100 software vendors to optimize apps for Intel’s NPUs. This includes collaborations with Microsoft for Copilot+ features and Adobe for AI-enhanced creative tools. In the concall, Tan announced new wins, including awards from partners like Connection for driving AI PC adoption.
Challenges remain. Adoption has been slower than hoped in some markets due to higher prices and the need for ecosystem maturity. An analyst questioned the ramp: “How do you accelerate beyond enthusiasts?” Tan replied, “We’re seeing 40% of new PCs shipping with Arm alternatives, but our x86 foundation gives us scale. Expect broader penetration in 2026 as costs come down.”
For 2026 guidance, Tan projected AI PC shipments to grow 50%, contributing to overall CCG revenue growth of 10-15%. This push is crucial as traditional PC sales stabilize post-pandemic, with AI as the key differentiator.
CEO Lip-Bu Tan’s Turnaround Plan: A Roadmap to Revival
At the heart of the Intel Q4 2025 concall was Tan’s update on his turnaround strategy, which he calls “Foundry First” with a cultural reset. Tan, known for his no-nonsense style, outlined three pillars: financial discipline, ecosystem revitalization, and refined AI focus.
First, financial discipline. Tan has slashed capex, divested non-core assets (like dropping plans to spin out the Network and Edge Group), and secured $7 billion in external funding from Nvidia ($5B) and SoftBank ($2B). This bolstered Intel’s balance sheet, reducing cash burn from $8.4 billion in the first three quarters to a more manageable level. “No more blank checks,” Tan emphasized. “Every investment must tie to customer commitments.”
Second, revitalizing the x86 ecosystem. Intel is doubling down on partnerships to counter Arm’s encroachment in PCs. This includes co-engineering with HP and others for AI PCs and pushing open standards for AI development. Tan also streamlined leadership, with key groups reporting directly to him for faster decisions.
Third, a refined AI strategy emphasizing energy efficiency. With data centers consuming massive power, Intel’s Gaudi 3 AI accelerators and Xeon processors focus on cost-effective inference. Tan announced Crescent Island, a 2026 GPU for efficient AI at the edge, positioning Intel against Nvidia’s dominance.
Tan addressed cultural changes: “We’re becoming faster, more agile. We’ve reduced layers, empowered teams, and fostered accountability.” He acknowledged past missteps, like delays in 10nm nodes, but stressed progress on 18A, with sampling underway and production set for H2 2026.
Analysts praised the plan but probed on execution risks, like yield issues or competition. Tan was optimistic: “2025 was stabilization; 2026 is acceleration. We’re on track for 60% gross margins by 2030.”
Analyst Reactions and Market Impact
Post-concall, Intel’s stock jumped 6% in after-hours trading, reflecting relief over beaten guidance and positive outlook. KeyBanc upgraded to Overweight with a $60 target, citing sold-out AI server CPUs and potential 10-15% price hikes.
However, some remain cautious. Morningstar noted ongoing cash burn as a concern, while Seeking Alpha highlighted downside risks if 18A ramps falter. Overall, sentiment is improving, with Tan’s credibility boosting confidence.
Looking Ahead: Intel’s 2026 Outlook and Beyond
Tan provided 2026 guidance: revenue $55-58 billion, with foundry growing 15% and AI contributing 20% of total. EPS is expected at $1.20 non-GAAP, with margins expanding to 42%. Key milestones include Panther Lake volume shipments, 14A development, and more external foundry deals.
Geopolitical factors loom, with U.S. subsidies (now equity via Trump’s administration) providing a buffer. Tan stressed diversification: “We’re building resilience across regions.”
In summary, theIntel Q4 2025 concall painted a picture of a company in transition—still facing headwinds but gaining momentum under Tan’s leadership. Foundry losses are easing, the AI PC push is accelerating, and the turnaround plan is taking root. For investors, it’s a story of patience and potential payoff in the AI era.
As Intel navigates this pivotal phase, one thing is clear: the chip wars are far from over, and Intel is fighting back with renewed vigor.
You must be logged in to post a comment.