The Semiconductor Market Just Crossed $110B in a Single Month. The Forecast Has Been Revised Up Twice Since January.

The Revision Is the Story

The number that matters is not $110.5 billion. It is the forecast trajectory. In February 2026, the SIA reported full-year 2025 sales of $791.7 billion and projected annual global sales would reach roughly $1 trillion in 2026. Four months later, SIA endorsed the WSTS Spring 2026 forecast, which now projects annual global sales will grow by 90% to $1.5 trillion in 2026. A $500 billion upward revision in a single forecast cycle signals that demand is arriving faster and in larger volumes than even the industry's own member companies anticipated at the start of the year.

What the Monthly Numbers Actually Show

Global semiconductor sales were $110.5 billion in April 2026, an increase of 11% compared to the March 2026 total of $99.5 billion and 93.9% more than the April 2025 total of $56.9 billion. April marks the first month to cross $100 billion in industry history.

The sequential acceleration matters as much as the year-over-year figure. Monthly sales in 2026 have compounded sharply: January came in at $82.5 billion, February at $88.8 billion, March at $99.5 billion. This is not a seasonal pattern but a demand ramp.

Global semiconductor sales increased on a month-to-month basis for the 14th consecutive month in April. Fourteen straight months of sequential growth eliminates any restocking narrative from the 2023-2024 inventory correction.

One technical caveat: monthly sales figures are compiled by the World Semiconductor Trade Statistics organization and represent a three-month moving average. The April print is therefore smoothed across February through April, which understates peak-month momentum at the margin.

Geography Reveals the Demand Source

Year-over-year sales in April were up in the Americas (115.8%), Asia Pacific/All Other (114.9%), China (78.6%), Europe (54.7%), and Japan (15.6%).

The Americas leading at 115.8% is a direct signal. US hyperscalers — the primary buyers of AI accelerators, high-bandwidth memory, and advanced packaging — are procuring at a rate that doubles the prior-year baseline. Asia Pacific reflects TSMC and Samsung fab loading plus regional cloud operator procurement. China's 78.6% growth despite export controls on leading-edge logic suggests continued strength in legacy-node and memory segments.

Japan at 15.6% stands apart. Japanese chipmakers concentrated in specialty materials, sensors, and mature-node analog are not at the center of the current cycle.

The $1.9 Trillion Signal

In 2027, global semiconductor sales are projected to exceed $1.9 trillion. That figure warrants skepticism.

The WSTS forecast aggregates member company inputs rather than applying independent econometric modeling. WSTS tabulates its semi-annual forecast by gathering input from an extensive group of global semiconductor companies. Member companies have clear incentives to project demand that justifies their capital expenditure plans and supply commitments. The February forecast said $1 trillion; the May forecast says $1.5 trillion for the same year. A 50% upward revision in four months should temper confidence in a 2027 projection made today.

The directional signal, however, is real. The SIA attributed the growth trajectory to increasing demand for AI infrastructure and accelerated computing platforms. AI model training and inference workloads are computationally discontinuous from prior enterprise cycles. A single large language model training run at frontier scale can consume more GPU-hours than an entire datacenter's prior baseline. Procurement teams at AWS, Microsoft, Google, and Meta are managing this operating reality.

The Revenue vs. Volume Distinction

High prices inflate revenue without proportional increases in unit volume or installed capacity. The 2025 full-year data supports this concern: global semiconductor sales hit $791.7 billion in 2025, an increase of 25.6% compared to the 2024 total of $630.5 billion. Memory, particularly high-bandwidth memory (HBM), has seen significant ASP inflation.

SK Hynix and Micron have both reported HBM allocation constraints. When a product is supply-constrained, revenue growth overstates capacity expansion. An operator procuring HBM for inference infrastructure should not read $110.5B in monthly sales as confirmation of readily available supply. It may instead confirm that existing supply is selling at elevated prices to committed buyers, with spot market access limited.

Where the Next Constraint Forms

Foundry capacity at leading nodes remains the conventional bottleneck. The more immediate risk, based on current demand, lies in substrate and advanced packaging. High-bandwidth memory requires interposer technology — CoWoS at TSMC, for instance — to connect memory stacks to compute dies. That packaging capacity is concentrated and has not scaled at the same rate as raw wafer starts.

Equipment vendors tooled for mature geometries face a different problem: demand rather than capacity limits them. If the $1.5 trillion projection is driven by AI silicon and HBM rather than broad-market recovery, chipmakers without AI-specific roadmaps will watch utilization rates stagnate while headline numbers climb.

Advanced node capacity sits overwhelmingly in Taiwan and South Korea — a structural risk no forecast cleanly captures. At current utilization, any disruption does not just create a supply shock but a multi-year recovery problem given the 3-5 year lead time to bring a new leading-edge fab online.

What to Watch

Q3 capacity utilization at TSMC and Samsung. If both hit ceiling simultaneously, spot pricing for advanced packaging and HBM interposers will diverge sharply from contract pricing. Their Q2 earnings calls in July will telegraph this.

The May and June WSTS monthly prints. If sequential growth slows from April's 11% pace, the $1.5 trillion annual projection becomes strained. A flat or negative June reading contradicts the narrative.

Supply commitment duration in hyperscaler earnings. AWS, Azure, and Google Cloud report in late July. Language around multi-year silicon commitments and capex guidance revisions will confirm whether demand is structural or whether customers are pulling forward orders ahead of tariff or export control changes.

Memory ASP trends as a leading indicator. If HBM spot prices soften while volume grows, new capacity is arriving and the packaging bottleneck is loosening. If prices hold or rise alongside growing volume, the constraint is durable.