The tale of Intel’s return literally involves a barge. A few years ago, a huge cold box that was too big to fit under highway overpasses close to Cincinnati was pushed up the Mississippi and Ohio Rivers by a tugboat. Eventually, workers loaded it onto a specialized trailer, and as the convoy crawled toward New Albany, Ohio, at eight miles per hour, cherry pickers moved traffic lights out of the way. A two-hour trip was extended to two weeks. That picture, which is laborious, costly, and slow, seems to be a true representation of everything Intel is currently attempting to accomplish.
With great fanfare, Ohio One, Intel’s proposed semiconductor mega-factory in Licking County, was revealed. The location was referred to as a “field of dreams” by President Biden in a State of the Union speech. The CEO of Intel at the time, Pat Gelsinger, predicted that the company might grow to be the biggest chip factory on the planet. The project received grants and loans totaling $8.5 billion from the federal government. When Intel was at its most ambitious, it was planning to invest $28 billion and have two fully operational factories by 2032. It was the kind of announcement that causes local officials to start daydreaming about job numbers and politicians to post videos on social media.
Then, slowly, reality began to set in. For the entire year 2024, Intel reported a net loss of $18.8 billion. With only $4.4 billion in mostly internal sales, its foundry division alone lost $2.5 billion in a single quarter. Ohio’s construction slowed. When the company announced in July 2025 that it was pulling back on the project, executives discreetly mentioned that Ohio One might be completely canceled if Intel’s next-generation manufacturing process, known as 14A, failed to draw in outside buyers. That was a sobering statement to read in a corporate filing for a project that was meant to establish American semiconductor independence.
Although Ohio is not the only state where Intel faces this fundamental issue, Ohio makes it extremely apparent. It takes a lot of money to run a foundry and make chips, not just design them. It took decades and hundreds of billions of dollars for TSMC, the massive Taiwanese company that currently controls the production of advanced chips, to become what it is today.
During a period of underinvestment in the 2010s, Intel lost its manufacturing advantage, lagged behind in critical processes, and saw competitors like AMD and Apple ship their chip orders to Taiwan. Rebuilding not only facilities but also trust is necessary to catch up, and in the semiconductor industry, trust is determined by delivered wafers rather than press conferences.
Whether Intel can completely close that gap is still up in the air. If the company’s 2027 14A process is successful, external clients may start placing orders. Recent signals are cautiously positive. The fact that Intel repurchased a 49% share in its Irish factory indicates that the balance sheet is at least improving. Since taking over as CEO in early 2025, Lip-Bu Tan has met with Jensen Huang of Nvidia, Apple, and even Elon Musk’s fledgling Terafab project. Nvidia declared in September that it would invest $5 billion in Intel and employ custom Intel CPUs. Just that news caused Intel’s stock to rise 23%.
Perhaps more than any market force, there is a perception in the industry that Intel is now firmly supported by the political wind. According to reports, Commerce Secretary Howard Lutnick has been pressuring big tech firms to collaborate with Intel. In 2025, the U.S. government paid $8.9 billion to acquire a 10% share in the business, fulfilling the remaining commitment under the CHIPS Act. It is difficult to disagree with Washington’s assessment that Intel is too strategically significant to fail in a world where TSMC produces more than 90% of advanced chips from a single island in the Pacific.
However, the issue of manufacturing execution is not resolved by government assistance. In terms of yield, reliability, and cost—the factors that really matter to Apple or Nvidia when they’re choosing where to manufacture a chip worth hundreds of dollars each—Intel still needs to demonstrate that its processes can compete with TSMC. The Ohio location won’t be completely operational until at least 2027, and both of the planned factories might not be operational until 2032. Over that time, chip technology can change significantly.
In some respects, what’s actually taking place in New Albany at the moment is amazing: enormous machinery is being delivered in pieces, and the construction site is home to one of the most costly industrial projects in American history. In a different way, it serves as a reminder of how far Intel strayed before attempting to turn things around. The barge arrived in Ohio. For a generation, American chipmaking will be defined by whether or not the factory fulfills its promises.
