Skip to content
News

Power Semiconductor Supplier Cites AI Data Centers as Demand Driver as August Revenue Rises

DataNews Editorial Team

Power semiconductor supplier PANJIT International reported August revenue of TWD 13.77 billion, up 29.82% year on year despite a 4.89% month-on-month decline. The company identifies AI data centers and AI servers among the demand drivers for its power-device portfolio.

Highlights
  • August revenue was TWD 13.77 billion, up 29.82% year on year and down 4.89% month on month.
  • AI-related applications represented about 11% of first-half revenue.
  • MOSFETs, used in relevant power architectures, accounted for 32% of first-half revenue and were the company’s largest product line.
  • The disclosure is an equipment-supply signal, not evidence of new data-center capacity or a named customer contract.

PANJIT reported August revenue of TWD 13.77 billion, representing year-on-year growth of 29.82% and a 4.89% decline from the previous month. Revenue for the first eight months reached TWD 102.01 billion, up 18.10% year on year.

The company identifies AI data centers, alongside automotive electronics and product-mix improvement, as contributors to business growth. AI-related applications accounted for about 11% of first-half revenue, according to the source.

Power devices including MOSFETs and diodes are used in server power supplies, power-conversion and protection systems, and cooling fans. The source notes that increasing GPU and ASIC power levels are raising data-center requirements for power efficiency and supply stability.

MOSFETs were PANJIT’s largest product line in the first half, accounting for 32% of revenue. First-half revenue was TWD 73.77 billion, up about 14% year on year, while second-quarter revenue was TWD 39.56 billion, up 15.6% quarter on quarter and 16.3% year on year.

For data-center infrastructure markets, the growing AI contribution to a power-component supplier’s sales could indicate a widening addressable market for efficient, reliable electrical components across the AI data-center supply chain. However, this single company result does not establish the size of that market, data-center commissioning rates, or purchasing volumes from any individual hyperscaler or other operator.

Why It Matters

The result is a corporate-level indicator that rising power requirements in AI servers may be translating into demand for power semiconductors used in data-center electrical infrastructure. It does not evidence new data-center capacity, a specific operator contract, or the scale of procurement by cloud providers.

Sources

This article was prepared by the DataNews editorial team based on the sources listed above.