Revenue across the global foundry industry grew 25% quarter-over-quarter in Q2, according to figures reported for the "Foundry 2.0" segment, as artificial intelligence demand pulled the entire semiconductor ecosystem upward rather than concentrating gains at a single vendor or process node.

The 25% sequential surge is the headline number, and its breadth is the story. In previous cycles, outsourced chip manufacturing tended to move unevenly: leading-edge logic raced ahead while mature nodes idled, or a single dominant foundry captured most of the upside. This quarter, the recovery spanned the wider Foundry 2.0 definition — a category that includes not only pure-play advanced foundries but also the broader base of contract manufacturers serving mature process geometries, specialty technologies, and discrete products.

AI is the engine. The buildout of AI computing infrastructure has created demand that radiates outward from advanced logic wafers into everything around them: memory-adjacent silicon, power management, analog components, interconnect, packaging services, and the legacy-node chips that populate server boards, networking equipment, and data-center power delivery. A 25% quarterly jump in aggregate foundry revenue signals that this radiating demand has now reached contract manufacturers across the stack, not just the companies printing the smallest transistors.

The scale of the move matters for capacity planning. Foundries price and allocate wafer capacity on quarterly cycles, and a one-quarter revenue expansion of this size typically reflects a combination of rising wafer shipments, richer product mix, and urgent pricing on tight capacity. For fabless customers, the practical consequence is straightforward: bargaining power has shifted back toward the manufacturers. Design houses that enjoyed aggressive price concessions and available slot capacity during the 2023 inventory correction now face a supplier base that is filling up.

The timing also marks a turning point in the industry's inventory cycle. Foundry revenue had spent much of the past two years working through excess stock that fabless customers accumulated during the pandemic-era shortage. A 25% sequential increase in Q2 indicates that depletion phase has given way to genuine restocking and new demand — driven not by channel speculation but by AI infrastructure orders that customers need fulfilled on committed schedules.

The composition of the recovery carries commercial weight. When demand growth is broad-based across the Foundry 2.0 universe, specialty and mature-node suppliers regain utilization leverage, which supports pricing across commodity process technologies. That dynamic contrasts sharply with the leading-edge-only booms of prior AI quarters, where gains accrued almost entirely to the manufacturers of the most advanced logic wafers. Equipment suppliers, materials producers, and OSAT (outsourced assembly and test) partners all benefit when the expansion is ecosystem-wide, since mature-node capacity additions require lithography, deposition, and metrology tools just as advanced fabs do.

For the competitive landscape, the breadth of the recovery reduces the risk of a two-tier market. If AI demand had remained confined to leading-edge logic, the industry could have split into a capacity-tight advanced segment and an oversupplied mature segment — a scenario that would have triggered price wars at older nodes even as advanced wafers commanded premiums. A 25% lift across the broader foundry base suggests the two tiers are rising together, at least for now.

Fabless customers will be watching utilization rates and lead times into the second half of the year. Historical patterns in the foundry business show that when quarterly revenue expands at this pace and AI infrastructure spending continues, capacity gets allocated through long-term agreements and prepayments rather than spot engagement — a structural shift that advantages large, well-capitalized customers and pressures smaller design houses into earlier commitments.

The quarterly figure also sets a demanding comparison for Q3. A 25% sequential jump creates a high base, and analysts will be looking at whether the momentum reflects a sustained demand step-change or a partial catch-up after a weak first quarter. What is not in dispute from the reported numbers is the direction: contract chip manufacturing is growing fast, and AI is the reason.

The forward question is durability. If AI infrastructure investment continues at its current pace, foundry utilization should tighten further through the second half, supporting pricing across both advanced and mature nodes and reinforcing the revenue momentum established in Q2.