Semiconductor Growth Imbalance, Sales to Top $2T by 2036
According to Deloitte’s latest 2026 Global Semiconductor Industry Outlook, global semiconductor sales are projected to reach a record $975 billion in 2026, driven largely by accelerating investment in artificial intelligence (AI) infrastructure. However, despite this strong growth, the report highlights emerging risks tied to the industry’s increasing concentration on AI-related segments and calls for more balanced, forward-looking strategies.
The report notes that the semiconductor market expanded by 22% in 2025 and is expected to grow by a further 26% in 2026. Even if growth moderates in the following years, annual revenue could still surpass $2 trillion by 2036. Beneath these record figures, however, lies a structural imbalance: while high-value AI chips account for nearly half of total industry revenue, they represent less than 0.2% of total unit shipments. At the same time, growth in chips used for automotive, PCs, smartphones, and non-data-center communications is comparatively slower.
AI demand is also reshaping the semiconductor supply chain. Memory revenue is forecast to reach $200 billion in 2026, accounting for about 25% of total industry sales. Strong demand for advanced memory technologies such as HBM3, HBM4, and DDR7 is tightening supply of mainstream memory like DDR4 and DDR5. Prices for these products surged roughly fourfold between September and November 2025 and are expected to rise further—by up to 50%—in the first half of 2026. This AI-driven competition for wafer fabrication and advanced packaging capacity is creating a “zero-sum” dynamic, introducing new systemic risks for industry players.
To support rapidly growing AI workloads—expected to increase three- to fourfold annually—chipmakers are accelerating the adoption of chiplet-based architectures to improve yield and energy efficiency. High-bandwidth memory (HBM) is increasingly integrated નજીક logic chips to enable data transfer speeds at the terabyte-per-second level. Meanwhile, co-packaged optics (CPO) and linear pluggable optics (LPO) are expected to see broader deployment in 2026, reducing power consumption by 30% to 50% while enhancing bandwidth efficiency.
Investment in AI networking infrastructure is projected to grow at a compound annual rate of 38% between 2024 and 2029. At the same time, strategic partnerships among AI developers, semiconductor companies, and cloud service providers are signaling the start of a new AI-driven capital cycle. Investments made in 2025 are likely to continue or accelerate into 2026, fostering an ecosystem where capital and computing resources circulate more dynamically. This trend may enable semiconductor companies to achieve deeper vertical integration across the AI data center stack through cyclical financing models.
