Over the past few months, I have frequently heard from tech and chip veterans that this AI revolution is the golden era they have waited decades for. At the core of this revolution is the semiconductor industry, which powers nearly every aspect of modern technology and, according to ASML, will be worth $1 trillion globally in revenue by 2030. “A single chip can unlock incredible possibilities. Without them, our economy and security are at risk,” highlights Bruce Andrews, chief government affairs officer at Intel.
Chips are indispensable for computational tasks and are likely more ubiquitous than you imagine. “From the moment you wake up, our products are at work,” says Ganesh Moorthy, CEO of Microchip. “Your climate control system, whether it is an air conditioner or a thermostat, likely contains our microcontrollers and analog products. Your alarm clock, home security system and even your electric shaver or toothbrush incorporate our semiconductors. As you make breakfast, our technology is present in your coffee machine, microwave and kitchen appliances,” adds Moorthy. As computational demands grow more complex, especially due to AI, the need for advanced chips increases.
The U.S. holds a key advantage in chips, in that “it has maintained leadership in design, with companies like Apple, NVIDIA and Google leading the way,” says Anirudh Devgan, president and CEO of Cadence. “If we lose our edge in design, no amount of manufacturing can compensate.” The design of these chips is the engine fueling innovation, while the growing convergence of semiconductor and system companies is transforming the industry. Companies like NVIDIA, once purely semiconductor players, are now regarded as system companies, while major system players like Apple and Google have built large in-house silicon teams. This dynamic means system companies are designing tailored chips for their specific needs, as seen in Tesla’s custom silicon driving its EV innovation. Michael Schmidt, director of the CHIPS Program Office, further adds that “a key advantage is that major buyers of leading-edge chips—companies like Apple, NVIDIA, and Qualcomm—are American, which bolsters the U.S. position.”
Yet, to capitalize on this advantage, the U.S. is now focusing again on manufacturing to create a pipeline from design all the way to production, “despite the prevailing notion that manufacturing was no longer viable in the U.S.,” underlines Thomas Sonderman, CEO of SkyWater Technology. “The push for ‘Made in America,’ which began under the previous administration, has accelerated, and we are at the forefront of bringing semiconductor manufacturing back to the U.S.,” adds Sonderman.
Playing Catch-up
Once producing nearly half of the world’s chips, the U.S. semiconductor industry now produces only 10 percent, relying heavily on East Asian imports, with TSMC—the Taiwanese giant—manufacturing the majority of advanced chips. The COVID-19 pandemic exposed the dangers of this dependency, with supply chain disruptions reverberating across industries. “Semiconductor shortages during the pandemic exposed their critical role, rallying bipartisan support for the CHIPS Act to secure national security and economic stability,” underlines John Neuffer, CEO of the Semiconductor Industry Association (SIA).
Indeed, efforts to rebuild domestic production are gaining momentum with the CHIPS and Science Act at the heart of this manufacturing push. It is the U.S. government’s largest-ever investment in semiconductor production, with $52 billion allocated to strengthen manufacturing, research, and supply chains. This historic funding has catalyzed transformative projects like Intel’s $100 billion expansion and TSMC’s $65 billion Arizona facility. But it is not as simple as bringing TSMC to the U.S. Fabrication labs (fabs) take years to build and as Jean-Christophe Eloy, CEO and president of Yole Group, adds: “The entire supply chain involves hundreds of companies with specific skills that are not easily relocated.”
Moreover, this reshoring may be made more difficult in 2025. Speaker Mike Johnson’s suggestions of rescinding CHIPS Act funding highlights the need for consistent bipartisan support. Without it, U.S. companies will struggle to compete with East Asia’s homegrown companies. As Ganesh Moorthy, president and CEO at Microchip, shares: “Over the years, semiconductor manufacturing shifted abroad due to various incentives provided by other countries. Bringing it back requires significant capital, often needing government assistance to reduce risks. Countries like Taiwan, Korea, Japan, Israel and now Europe and India have historically offered such incentives.”
Bruce Andrews of Intel adds: “We do not just compete with companies, but with entire countries.”