In the vast Phoenix Valley desert, Intel has invested over $20 billion into a colossal four-story manufacturing plant, the cornerstone of the struggling chipmaker’s ambitious recovery plan.
This facility, known as Fab 52, is where Intel is pioneering a groundbreaking new manufacturing process designed to create more powerful and energy-efficient computer chips. For the first time in almost a decade, Intel is utilizing advanced tools from ASML, the Dutch lithography machine manufacturer, to produce its cutting-edge semiconductors right here in the United States.
During a recent visit to the factory, two of these impressive $250 million machines sat largely inactive, with ASML engineers in pristine white clean suits diligently working on one. Nearby, two empty spaces, each the size of a trailer, hinted at Intel’s future expansion hopes.
Intel has been showcasing Fab 52 to numerous prospective clients in Chandler, Arizona, hoping to convince them to manufacture their chips there. However, industry analysts suggest that most chip companies are waiting to see if Intel can first successfully produce its own advanced chips before entrusting it with the production of semiconductors for smartphones, artificial intelligence systems, and other critical technologies.
The intense focus on this new facility highlights the critical importance of this moment for Intel’s future. Once a shining example of Silicon Valley innovation, the company has fallen behind rivals in both chip manufacturing and design. Intel was surpassed by Taiwan Semiconductor Manufacturing Company (TSMC) after failing to adopt ASML technology effectively, and its chips were even dropped from Apple laptops due to performance and battery life issues.
As sales plummeted, Intel saw a rapid succession of leaders. In March, Lip-Bu Tan was appointed its third chief executive in five years. He inherited a precarious financial situation and an ambitious plan: to roll out five new production processes in just four years, all while transforming Intel into a contract manufacturer for other chip designers.
Mr. Tan stabilized the company’s finances in August when the Trump administration injected $8.9 billion in exchange for a 10 percent stake. This deal marked one of the largest U.S. government investments in a company since the 2008 financial crisis.
The Trump administration’s investment came from funds allocated under the CHIPS and Science Act, a federal initiative passed in 2022 to provide grants aimed at revitalizing U.S. semiconductor manufacturing.
Back in 2021, while the CHIPS Act was still under consideration, Intel committed $32 billion to build two new factories, Fab 52 and Fab 62, at its Ocotillo campus in Arizona. This project later received federal funding support.
Now, Mr. Tan’s primary objective is to execute the second phase of Intel’s strategy: a comprehensive technical resurgence.
At a recent event attended by hundreds of technology analysts and journalists, Sachin Katti, Intel’s chief technology officer, underscored the high stakes. He described the new manufacturing process, dubbed ’18a,’ and the ‘Panther Lake’ computer chip as “foundational to our future.”
“We are, of course, making two really big bets here,” he stated, emphasizing that these manufacturing efforts are “mission critical, not just for Intel but for our country.”
According to Ben Bajarin, principal analyst at Creative Strategies, Intel once dominated the chip industry by strategically adopting new technologies. Typically, the company would develop a new manufacturing process to etch billions of minuscule transistors onto silicon, then perfect it on an older generation of chips before applying it to newly designed components.
However, this time, Intel is attempting both manufacturing and design breakthroughs simultaneously. The 18a process utilizes ASML tools to create chips with stacked transistors, drastically reducing their footprint. It also reroutes power to the backside of the chip, avoiding previous clashes with data flow at the top. Intel claims this results in a denser, more energy-efficient chip capable of layering 10,000 sheets of silicon in a stack thinner than a single sheet of paper. The company projects its Panther Lake series will power laptops running advanced AI systems and offer all-day computing, with shipments beginning early next year.
Intel’s confidence in this process is so strong that it’s shifting production of its newest chips back to its own factories, rather than relying on TSMC, which it previously turned to after earlier processes faltered. Yet, these simultaneous changes introduce more potential pitfalls, warned Mr. Bajarin, especially in a process where a single dust particle can destroy a chip and lead to significant financial losses.
“The product has to be good, and the process has to be good,” Mr. Bajarin emphasized. If these conditions are met, he believes, the industry will acknowledge that Intel is once again operating at full capacity.
Intel’s success, however, is far from assured. Late last year, some customers were informed that Intel’s 18a process was lagging behind competitors. At that time, TSMC was flawlessly producing 30 percent of its leading-edge 2-nanometer chips, while Intel’s 18a process yielded fewer than 10 percent flawless chips.
During last month’s two-day event, Intel executives refrained from commenting on the current percentage of flawlessly manufactured chips at Fab 52. Analysts anticipate that the true yield rates will only become clear once the company releases its financial reports next year.
By repatriating some production from TSMC, Intel could realize significant cost savings and boost profits, noted Patrick Moorhead, founder of Moor Insights & Strategy, a tech research firm. This move also serves as a crucial demonstration for major chip producers like Nvidia and Apple, who currently rely exclusively on TSMC, that Intel is a viable alternative for manufacturing their semiconductors.
Mr. Moorhead pointed out that Intel does not produce enough of its own chips to fully utilize the Fab 52 plant. But if it can convince other tech industry players that its new process is reliable and efficient, it could invest in more ASML machines and maximize its factory’s capacity.
“It’s like a shopping center that needs an anchor tenant,” Mr. Moorhead mused. “If it can find one, it would be game on.”