Intel's High-Stakes Gambit: Inside the Turnaround to Conquer AI

BALTIMORE, MD – June 17, 2026 – When Intel CEO Lip-Bu Tan sat down for a conversation with T. Rowe Price’s Chief Investment Officer, Eric Veiel, for the asset manager’s "The Angle" podcast, the discussion was more than just another C-suite interview. It was a public articulation of one of the most audacious and closely watched turnaround efforts in modern technology: Intel's fight to reinvent itself for the age of artificial intelligence. The conversation, framed around the infrastructure demands of AI, offered a window into a company attempting to shed decades of inertia and reclaim its throne.

For leaders and investors alike, the core question is not about the vision, but the execution. While rivals like NVIDIA and AMD have captured the narrative and market share in the AI boom, Intel is now making a forceful, multi-pronged counterattack. Under Tan, who took the helm in March 2025, the semiconductor giant is betting its future on a radical restructuring, a high-risk foundry strategy, and a renewed focus on the fundamental building blocks of compute—from CPUs and GPUs to the advanced packaging that ties them all together. The stakes are immense, but early signs suggest the sleeping giant may be stirring.

The Tan Doctrine: Speed, Execution, and a Flatter Hierarchy

The most immediate changes at Intel under Lip-Bu Tan have been cultural and structural. Inheriting a company criticized for its bureaucratic sluggishness, Tan has implemented what insiders call a "speed-of-light" operating model. This is not just corporate jargon; it represents a fundamental rewiring of the company’s operational DNA. The management hierarchy has been aggressively streamlined, collapsing from twelve layers down to just five. This move is designed to accelerate decision-making and increase accountability, putting engineering leaders in direct contact with the CEO.

This overhaul has been accompanied by a significant and painful restructuring, reportedly involving a 15% reduction in staff, primarily targeting management roles. The goal is to create a leaner, more agile organization that prioritizes engineering prowess and customer focus over internal process. In the T. Rowe Price discussion, Tan emphasized the importance of listening to customers, strengthening execution, and moving with greater speed—a direct acknowledgment of past failings that saw Intel lose ground in process technology and product roadmaps. This ruthless focus on operational excellence is the bedrock of the entire turnaround strategy. If Intel can't build and ship on time, its grander ambitions in AI and foundry services are moot.

The Foundry Bet: A Trillion-Dollar Challenge to TSMC

At the heart of Intel’s reinvention is Intel Foundry Services (IFS), an ambitious and capital-intensive plan to become a world-leading contract chip manufacturer, directly challenging the dominance of Taiwan's TSMC. This strategy, initiated under former CEO Pat Gelsinger, is now being accelerated by Tan. The crown jewel of this effort is the Intel 18A process node, which promises to restore the company’s technological leadership. The node officially entered risk production on schedule in June 2026 with its performance-enhanced 18A-P variant, a critical milestone demonstrating that the firm's manufacturing roadmap is finally getting back on track.

The numbers show both the promise and the peril of this bet. In the first quarter of 2026, the foundry division’s revenue grew an impressive 16% to $5.4 billion. However, it also posted a substantial operating loss of $2.5 billion in the preceding quarter, underscoring the massive upfront investment required. Yet, momentum is building. The company has secured a reported preliminary agreement to manufacture chips for Apple and a $5 billion investment from NVIDIA, which plans to use Intel's advanced packaging services. Furthermore, a multiyear collaboration will see Google Cloud deploy Intel's new Xeon 6 processors.

This push is not just a business strategy; it's a geopolitical one. With the backing of the U.S. CHIPS Act, which provided Intel with $5.7 billion in government funding, IFS is being positioned as a crucial "Western manufacturing platform." This aims to de-risk the global semiconductor supply chain, which is heavily concentrated in Asia. As Eric Veiel of T. Rowe Price noted, understanding Intel's role in rebuilding "foundry services" is a compelling theme for investors looking at "the forces shaping the next phase of technology leadership."

Battling on All Fronts: The Competitive AI Landscape

While the foundry bet matures, Intel is simultaneously fighting a fierce battle in the product arena. For years, NVIDIA has been the undisputed king of AI, its GPUs and CUDA software platform becoming the industry standard for training large language models. The company holds a near-monopoly on the AI training market, with a staggering 95% share of the gaming GPU market at the end of 2025. Meanwhile, AMD has mounted a formidable challenge, clawing its way to nearly 46.2% revenue share in the lucrative x86 server CPU market in Q1 2026, directly threatening Intel's traditional stronghold.

Intel’s response is a pincer movement. On one front, it is reasserting the importance of the CPU in the AI era. While GPUs excel at parallel processing for training, the rise of "agentic AI"—complex, multi-step AI applications—requires the orchestration, low latency, and single-thread performance where CPUs shine. Intel's new Xeon 6 processors, built on the cutting-edge Intel 18A process, are aimed squarely at this burgeoning market for AI inference and orchestration. The company still commands over two-thirds of the server CPU market, providing a massive foundation to build upon.

On the other front, Intel is not ceding the accelerator market. It is preparing to launch its Crescent Island data center GPU in 2026, targeting the energy-efficient AI inference space rather than directly competing with NVIDIA's training dominance. By focusing on performance-per-watt and offering a comprehensive portfolio of CPUs, GPUs, and custom accelerators, Intel hopes to provide a more holistic and cost-effective solution for data centers grappling with the immense power consumption and cost of AI infrastructure.

A Turnaround Taking Root

The combination of radical internal restructuring, a disciplined manufacturing roadmap, and a pragmatic product strategy appears to be gaining traction. The company’s first-quarter 2026 earnings beat analyst expectations, with the Data Center and AI division growing a healthy 22% year-over-year. This has been reflected in its stock performance, which has seen a remarkable recovery over the past year.

The conversation between Intel's CEO and T. Rowe Price's CIO was not a declaration of victory, but a clear signal of intent. It highlights a growing recognition among sophisticated investors that the AI revolution will be built on a complex foundation of hardware, where no single player may have all the answers. Intel's path is fraught with risk, and its competitors are relentless. But for the first time in years, the company is executing on a clear and coherent plan, transforming itself from a legacy giant into a formidable contender for the future of compute.