As an Amazon Associate, we earn from qualifying purchases. TweakTown may also earn commissions from other affiliate partners at no extra cost to you.
TSMC currently has 24 semiconductor factories under construction across Taiwan, the United States, Germany, and Japan, but the company has a shortage of fab workers that could slow things down.

In a new report from DigiTimes, we're hearing that almost all of the teams that TSMC has in Taiwan that can be mobilized, have been, and that hiring locals in Taiwan has caused issues to get worse across its global supply chain. TSMC had 83,825 employees at the end of 2025, which is a huge increase from the 51,000 staffers it employed at the end of 2019.
However, turn over of TSMC employees has reportedly been high enough that it has pushed the company to increase salaries and bonuses, with overseas operations being a challenge to figure out local work culture. Japan is very similar to Taiwan, but there aren't many workers to fill positions, while the shortage of workers in the US has made it hard for TSMC to deal with America's "work-life balance" idea.
TSMC is reportedly seeing around 2000 to 3000 people leave the company every year, but with the above issues along with a declining birth rate and young people's unwillingness to work in the semiconductor manufacturing industry with long working hours have caused TSMC to be quite worried, reports DigiTimes.
DigiTimes adds that the US semiconductor manufacturing industry is "in a slump" and that rumors in the industry see many employees leaving Intel, GlobalFoundries, and Texas Instruments to resign from their positions and join TSMC. However, there are large differences in corporate culture between Taiwan and the United States, which has caused conflicts between Taiwanese and foreign employees of TSMC in terms of salary and work content.
All of this paints a hopefully not-too-troubling issue for TSMC as the leading semiconductor manufacturer, with 24 new fabs in construction right now and a workforce that is dwindling... robotics and AI will most likely be used (and are being used right now) to fill those voids.