AI boom squeezes Taiwan's traditional industries with rising talent and capital costs

Taiwan is rapidly accelerating its pace amid the global artificial intelligence (AI) wave. Since 2023, the government has successively rolled out AI industry development blueprints and established R&D subsidies and industrial funds exceeding NT$100 billion, attracting a surge of domestic and foreign capital. Major semiconductor and information technology companies have successively expanded cloud computing data centers, invested in deep learning chips and automation platforms, with related investments surpassing NT$50 billion last year. This convergence of capital and policy has driven the scale of Taiwan's AI industry to grow by over 40% in just two years, creating a distinct industrial heating effect and pushing the capital market's enthusiasm for AI-related stocks to record highs.

At the same time, however, traditional industries such as textiles, steel, food processing, and conventional manufacturing have felt the pressure of resources being "crowded out." Experts point out that the AI industry's soaring demand for high-end R&D talent has caused salary levels for critical positions such as engineers, data scientists, and system architects to rise by nearly 30% over the past year. Many technical personnel who previously worked in traditional industries have shifted to high-compensation AI startups or large tech companies. On the capital front, venture capital and bank loan allocations have also tilted heavily toward AI-related projects, and traditional enterprises face higher thresholds and longer review cycles when vying for new investments, leading to rising capital costs.

Changes in energy demand have likewise exacerbated the plight of traditional industries. The power consumption of AI computing centers and large data centers is far higher than that of ordinary manufacturing plants. In the first half of 2024, Taiwan Power Company adjusted its electricity rate structure due to increased AI-related power demand, raising the base rate for industrial electricity. For steel and chemical plants reliant on high energy consumption, rising costs have directly eroded gross margins. Meanwhile, fluctuations in energy prices have also made these industries more risk-averse when formulating long-term investment plans. As a result, the operating costs of traditional industries continue to climb, yet it remains difficult to offset these through product upgrades or price pass-throughs, resulting in sluggish revenue growth.

On the demand side, the rapid application of AI technology has altered consumer and corporate procurement behaviors. The proliferation of smart manufacturing, the Internet of Things, and automation solutions has replaced many manufacturing processes that previously relied on manual labor, showing a downward trend in demand for low-value-added traditional products. Taking the textile industry as an example, AI-driven rapid design and customized services have made international brands more inclined to cooperate directly with technology suppliers, eroding the market share of local small and medium-sized enterprises. This shift in demand structure subjects traditional industries to dual pressures on the revenue front: rising costs and shrinking markets are fermenting simultaneously, forming a classic K-shaped economic polarization.

Faced with this increasingly pronounced industrial divide, calls for policy-level adjustments have begun to emerge. The Ministry of Economic Affairs and the Industrial Technology Research Institute have proposed "industrial upgrade and transformation" programs to encourage traditional enterprises to adopt AI-assisted production management systems and smart logistics to reduce labor costs and enhance production flexibility. Meanwhile, the Ministry of Labor plans to establish retraining subsidies to help technical talent absorbed by AI return to traditional sectors, cultivating cross-disciplinary skills such as digitization and machinery maintenance. If a balanced allocation can be achieved across the three major factors of capital, talent, and energy, it may mitigate the widening trend of the K-shaped economy and allow traditional industries to recapture a portion of the benefits from the technological dividend.