Once a rival to TSMC at the industry’s front line, UMC has redirected its strategy toward specialized chipmaking.

When United Microelectronics Corporation (UMC) was founded in 1980, it carried a mission that would help define Taiwan’s place in the global technology supply chain.

Spun off from the state-backed Industrial Technology Research Institute (ITRI), UMC emerged in 1980 as Taiwan’s first-ever semiconductor company — a bold move for an island that at the time had no chip fabrication capability of its own. The global stage was then dominated by entrenched U.S. giants like Intel and Texas Instruments and powerful Japanese electronics makers such as NEC and Toshiba.

Against this backdrop, UMC began life as an integrated device manufacturer (IDM), bringing chip design, wafer fabrication, and packaging together under one roof — a model that allowed it to control every stage of production while Taiwan’s fledgling semiconductor sector found its footing.

“In the early days, we had to do everything, because there was no one else in Taiwan to do it — we designed the chips, made them, packaged them, and built the teams from scratch,” says Kuo Linwu, vice president of public affairs and corporate communication.

But soon, the semiconductor landscape began to change. In 1987, the establishment of Taiwan Semiconductor Manufacturing Co. (TSMC) introduced the pure-play foundry model — manufacturing chips solely for external design houses, with no competing in-house product lines.

TSMC’s approach quickly gained traction, especially as a wave of fabless integrated circuit (IC) design firms emerged in Taiwan and abroad. By the early 1990s, many of these new design houses were becoming UMC customers, even as UMC continued to operate as an IDM, putting the company in a strategic dilemma.

“We could see the industry separating,” says Kuo. “Design companies were growing, and they wanted partners, not competitors. In order not to compete with our customers, we made the decision in 1995 to become a pure-play foundry.”

For a time, UMC competed directly with TSMC in leading technology. “We were two peers, racing for the same technology milestones, almost until the year 2000,” Kuo says.

The internal shift in culture and operations was significant, says Peter Tou, senior division director of corporate communication. “Becoming a pure-play foundry meant changing how we thought about everything,” he notes. “We had to put the customer’s roadmap at the center, build flexibility into every process, and align our R&D and operations so we could handle dozens of different products on the same line.”

However, the economics of the race became punishing after the 2000s. “It’s a ‘winner takes all’ situation, where the leading company would capture the lion’s share of the market and resources, which led UMC to rethink its strategy — specialty technologies,” Kuo adds.

The beauty of specialty technologies

Over time, UMC concentrated on mature yet technologically demanding processes where it could lead in both yield and performance. Tou frames it with an athletic metaphor:

“Imagine a world-class sprinter who realizes they can’t win the 100 meters anymore,” he says. “That speed can still win gold medals — in the long jump, the high jump, or as a wide receiver in football. We use our manufacturing speed and precision for specialized events, not just the straight sprint. That’s what specialty technology is for us.”

This strategy is evident today in UMC’s commanding position in 22-nanometer display driver ICs — critical components for smartphones, tablets, and high-resolution displays — where it holds about 70% of the global market. These chips must deliver precise voltage control to millions of pixels, and process improvements can directly affect screen quality, refresh rates, and power efficiency.

Similarly, UMC is a leader in 55-nanometer power management ICs, which regulate voltage and power flow in devices from laptops to automotive systems. As electrification spreads through industries such as transportation and industrial automation, demand for these high-reliability chips is expected to grow steadily.

“In specialty technologies, the R&D is just as intense as it is for the smallest nodes — sometimes more — because you have to co-develop processes with your customer, qualify them for high reliability, and adapt them to very specific applications,” Tou says. “That collaboration creates a relationship that is hard for low-cost competitors to break.”

According to industry tracker TrendForce, UMC is now Taiwan’s second-largest and the world’s fourth-largest contract foundry by revenue, with more than 400,000 12-inch-equivalent wafers produced per month across 12 fabs in Taiwan, Singapore, Japan, and China.

Changing business model

In an industry where process technology is usually kept under tight wraps, a new collaboration between Intel and UMC is challenging convention. In 2024, the Taiwanese chipmaker announced it would team up with Intel to co-develop a 12-nanometer FinFET platform for production at the latter’s Ocotillo Technology Fabrication plant in Arizona, targeted for 2027.

The agreement is unusual in the foundry industry, where manufacturers typically guard process technologies closely. It reflects both companies’ strategic needs: Intel has advanced process R&D but less experience in high-mix, multi-customer manufacturing, while UMC looks to advance beyond 14 nanometers without absorbing the full cost of node development.

“They have the technology, and we have the manufacturing know-how, flexibility and customer-facing expertise,” says Tou. “This is a true synergy.”

For UMC, the deal delivers two critical advantages: faster time-to-market for a new node and a U.S.-based manufacturing option for customers seeking diversification.

For Intel, it offers a way to broaden its foundry services portfolio, and transform its modus operandi of tailoring production to a wide range of customers.

Industry observers note that such collaborations could become more common as capital costs rise and governments push for domestic semiconductor production. In this case, the Arizona location aligns with U.S. policy priorities, giving UMC a foothold in the North American market without building a multibillion-dollar greenfield fab. “It is a valuable investment for UMC where we can secure manufacturing in the U.S. and support its supply chain,” adds Kuo. 

Pinpointing potential

UMC is moving ahead with this and other new projects amid a major sectoral challenge. The global semiconductor industry, already stretched by surging demand for advanced electronics, is bracing for an even sharper shortfall. Deloitte projects more than one million additional skilled workers will be needed by 2030 to staff fabrication plants, design centers, and operations hubs. That gap includes about 67,000 engineers in the United States, more than 100,000 professionals in Europe and over 200,000 experts in the Asia-Pacific region.

Taiwan, one of the world’s most critical chipmaking hubs, is feeling the strain. The Ministry of Labor counted roughly 34,000 open positions in key semiconductor roles in May.

To attract students in Taiwan, UMC partners with universities and semiconductor institutes to give students hands-on access to donated manufacturing equipment and sponsors science education programs to inspire future engineers.

“We work closely with the semiconductor schools, donate equipment, and send our engineers to lecture,” says Tou. “If students can see and touch the tools they will be using in their careers, they are much better prepared for the real work.”

The company is also working to refresh its employer brand in Singapore, where it has operated for two decades but historically did little promotion. “We realized that after 20 years, the initial excitement had faded,” Tou says. “We needed to show people that semiconductor manufacturing today is not a dark, noisy factory floor. It’s cleanrooms, cutting-edge R&D, flexible work, and a culture that values work-life balance.”

Facilities are part of the attraction. “In Singapore we have an indoor basketball court — and given the land costs there, that’s a statement about how much we value our people’s wellbeing,” Tou says.

In Taiwan, many employees take yoga classes or use the gym at lunch. “We want people to know that this is a place where you can build a career and have a life,” Kuo says.

Systems for success

Technology itself has become part of UMC’s recruitment strategy. Its digitalization drive not only boosts efficiency but also makes engineering roles more appealing. “We are introducing AI into our manufacturing to remove tedious, repetitive tasks from engineers’ workloads,” Tou explains. “That means they can focus on solving problems and innovating, which is far more appealing to top talent.”

While UMC does not target AI server processors, it is preparing for growth in edge AI devices and infrastructure, where its strengths in power management and data transfer can reduce latency and energy consumption. “Edge AI will need more than just computing power — it will need efficient ways to move data and manage power,” says Kuo.

Tou emphasizes the link between innovation and environmental responsibility, noting that sustainability has become a key differentiator for both talent recruitment and customer loyalty. “Every AI query uses energy and water somewhere,” he says. “By making AI more efficient at the edge, we’re helping to reduce that footprint.”

In water conservation, UMC recycles each drop 3.5 to 4 times, earning an “A” from CDP. It began reducing fluorinated greenhouse gases in the late 1990s and will open its Circular Economy and Recycling Innovation Center in Tainan in Q4 2025, turning waste materials into valuable assets, such as industrial sludge into fluorite, as an accelerant for steelmaking and turning copper sulfate liquid waste into copper tubes for resale.

“We see environmental protection not as a cost center, but as something that can generate economic value,” Tou says. “If you can recycle materials into something useful and even profitable, you create a strong incentive to do more.”

Brave new geopolitical world

While UMC’s technology leadership and global customer base remain central to its operations, the company’s senior leadership acknowledges that geopolitics now plays a far greater role in semiconductor business strategy than it once did.

“In the past, semiconductor companies didn’t really need to talk to government, but that has changed completely,” says Kuo. “Today, with U.S.-China competition, there’s export control policy, tariffs, and evolving trade rules. There’s a lot of dynamics coming from outside the company that you cannot ignore.”

This fact became obvious just days before reciprocal U.S. tariffs on Taiwan took effect on August 7, when U.S. President Donald Trump announced plans to impose a 100% tariff on imported chips. The United States imported more than US$60 billion worth of semiconductors in 2024, according to Politico, with Taiwan as the top supplier at about US$12 billion. The potential impact on the island is being closely watched.

The administration’s plan includes a key exemption for companies already manufacturing in the United States or committed to doing so — a provision that puts TSMC in a favorable position. With a US$165 billion U.S. investment pledge, TSMC is expected to avoid most of the blow, shielding major clients like Nvidia from higher costs on chips produced domestically.

This kind of policy uncertainty reinforces the need for semiconductor companies to expand their capabilities beyond technology and manufacturing. “When policy is shifting, you have to watch the direction carefully, prepare for it, and do real risk management,” Kuo says. “It’s no longer enough to just have the best fabs — you need the ability to understand and respond to policy, to work with allies, and to make sure your company’s interests are represented.”

Kuo’s own move to UMC from a Taiwanese government agency reflects a broader industry trend toward more targeted government engagement. “We have to anticipate the impact, manage the risks, and sometimes offer our perspective so that government understands the industry’s needs,” he says. That communication flows both ways. “Today it’s not unusual for the Taiwan government to call and ask how we see a particular policy, what challenges we anticipate, and what recommendations we have,” Kuo says. “There’s a recognition that industry and government need to work together if Taiwan is to remain competitive.”