Intel’s Balancing Act: Can It Be Both a Chipmaker and a Foundry?

Intel has a new CEO. Lip-Bu Tan is an accomplished technology leader with deep semiconductor industry experience. Will he take a bold new path and pivot Intel to the next?  Intel is still a behemoth. It has ruled the chip world in an almost unipolar market for a long time but found itself on a slippery slope over the last few quarters. AMD, once seen as an also-ran and on the brink in 2014, made a dramatic turnaround when Lisa Su took over the reins of AMD as President and CEO in 2014. She turned the headwinds into tailwinds and primed the company for the AI age.

Missing the Inflection Points

Did Intel miss the AI bus, like Microsoft missed the mobile bandwagon decades ago, despite having an early lead in handheld OS with Windows CE, which later became Windows Mobile in 2003? Microsoft did not see the mobile handset revolution coming, nor did it anticipate how iPhone will thrive along with operator subsidies and an app store ecosystem. Windows Mobile OS was functional at that time, but in 2007, when the iPhone was launched, then-Microsoft CEO Steve Ballmer ridiculed it, claiming it would never succeed. He also mocked the lack of a physical keyboard. The rest is history.

Every tech company has its highs and lows, and there comes a time in the evolution they get trapped in hubris, failing to foresee the distant future. Today, Intel is at a critical inflection point, where it needs radical innovations tuned to the AI age that is raging and impacting every business.

Anyone taking an honest view has to admit that the new CEO does not have a magic wand. He has to ‘do’ and ‘undo’ many things. A sort of reengineering of Intel’s hardwired DNA is going to be a challenging act. Tan said in a press statement, “I am honored to join Intel as CEO. I have tremendous respect and admiration for this iconic company, and I see significant opportunities to remake our business in ways that serve our customers better and create value for our shareholders.”

If one looks at the statement closely, the word ‘remake’ carries significant weight. This is where the doing and undoing part will manifest in all likelihood. In a multipolar chip industry now, being just an IDM (Integrated Device Manufacturer), designing and manufacturing its processors, is not enough. Let’s look at some of the strategic shifts in the making of Intel 2.0.

A Hybrid Chipmaker: IDM + Foundry

Around this time last year, in 2024, then-Intel CEO Pat Gelsinger created a very elaborate, well-meshed foundry strategy for Intel. Initially, Intel launched its Intel Foundry Services (IFS) in 2021, but the 2024 foundry strategy gave it more muscle. Gelsinger said at that time, “Intel’s differentiated position as both a world-class semiconductor manufacturer and a fabless technology leader creates significant opportunities to drive long-term sustainable growth across these two complementary businesses. Implementing this new model marks a key achievement in our IDM 2.0 transformation as we hone our execution engine, stand up the industry’s first and only systems foundry with geographically diverse leading-edge manufacturing capacity, and advance our mission to bring AI Everywhere.”

But Gelsinger's exit, a few months back,  was mostly attributed to that strategy not taking off as expected, while competition took the lead—with Nvidia dominating the GPU market, followed by AMD. It’s a moment of reckoning for Intel, for long, warming in the Wintel (Windows + Intel = CPU supremacy) comfort zone, while Nvidia, with early gains and extreme computing performance in the gaming industry, quickly morphed into enterprise-grade GPUs tuned for massive information processing for AI in real time, with Tensor cores (Optimized for AI acceleration), the much-needed computing requirement for GenAI.

A classic case of how Nvidia seamlessly transitioned from gaming GPUs to Generative AI needs, while  both Intel and AMD were polarized on CPUs—a strategic oversight.

Challenges Ahead

Intel might not sunset its ‘Foundry’ strategy, as much of the work has already gone into it. But how will that strategy pan out? Here are the key questions:

• If Intel starts making chips, it has to take TSMC head-on. But TSMC has already primed for AI. How long will it take Intel to bring AI chips to market? Does its fabs have the capabilities to bring chips that can square up to TSMC?
• How will the new CEO create an operational firewall between the foundry business and its processors? How will that pan out?
• Intel has long been mired in a product-first mindset. It must adopt a customer-first approach in the foundry business.
• Can Tan bake agility into Intel’s DNA so that it can become a nimble organization, quickly embracing market trends and turning them into real offerings?

Opportunities Ahead

Intel is not just a company; it’s an institution that has thrived on its co-founder Andy Grove’s mindset: “Only the Paranoid Survive.” Let’s look at some tailwinds, given its legacy.

• Intel has undisputed capabilities in manufacturing and has 15 wafer fabs in production globally at 10 locations. This puts it in a unique position to reimagine the fabs for more innovative product outcomes.
• The U.S. President’s push for domestic semiconductor production gives Intel a strategic advantage over TSMC (Taiwan) and Samsung (South Korea). But the caveat here is that Intel needs investment for its diversification. Recently President Trump indicating he might sunset the CHIPS and Science Act ( a $52.7 billion U.S. government initiative to boost domestic semiconductor manufacturing, research, and supply chain resilience) signed by former U.S President Joe Biden in August 2022, this does not bode well for cash-strapped U.S chip manufacturers. They may need to offset costs through higher export duties on their products.
• If Intel can leverage its foundry advantage and develop AI-focused chips, it might still make inroads into the data center GPU market.

Gearing up for a New Dawn?

Clearly, Intel needs to sunset many past-generation strategies to forge a new path. The road ahead has many twists and turns, and it has a lot of catching up to do to even appear in the rearview mirrors of leading GPU makers, who are far ahead of Intel on the road. No doubt, it’s a high-stakes gamble. Can Intel become the ‘TSMC of the West’ while maintaining its native processor business and the foundry, or will it falter in both? Can it pull off this delicate balancing act?

Nevertheless, Intel is at a pivotal point in its evolution. As H.G. Wells said, “Adapt or perish, now as ever, is nature’s inexorable imperative.”