ASE's LEAP: From Outsourced Capacity to Bottleneck King in the AI Packaging Era

Q2 2026 results confirm a strategic pivot—LEAP revenue to double in 2027 as ASE positions itself as the critical hardware infrastructure bottleneck for AI.

3711.TW · Earnings Call · 2026-07-30

The Pivot to Hardware Infrastructure

In its Q2 2026 earnings call, ASE Technology Holdings signaled a decisive strategic shift. For years, ASE was a bellwether of the semiconductor supply chain, but the conversation now centers on a new vocabulary: AI infrastructure as a bottleneck, LEAP services as the growth engine, and a capacity expansion that will consume billions of dollars. COO Dr. Tien Wu framed it starkly:

Hardware infrastructure is a bottleneck. With AI, the hardware requirement is new, insatiable and more complicated and more complex. And today, there are very few manufacturers capable of producing hardware. Therefore, it is the bottleneck today for capacity, for automation and more importantly, from an innovation perspective.
Tien Wu, Chief Operating Officer · 2026-07-30

This is not mere rhetoric. The numbers back it up. Consolidated Q2 revenue grew 27% YoY to TWD 191.1 billion, with ATM revenue up 36% to a record TWD 126.1 billion. Gross margin for ATM hit 27.3%, up 5.4 points annually, driven by operating leverage and a favorable mix shift toward LEAP services. The company raised its full-year ATM growth guidance from 13% to 20%—then to 35% for the year—and now expects LEAP revenue to exceed the earlier $3.5 billion target, with the explicit goal of doubling LEAP revenue in 2027. CFO Joseph Tung confirmed: “We are aiming to double our LEAP revenue in 2027.” — Joseph Tung, Chief Financial Officer · 2026-07-30 The strategic rationale is clear: ASE is moving up the value chain, positioning itself as an indispensable provider of advanced packaging and testing for AI accelerators, CPUs, and ASICs. As Dr. Wu put it, "Packaging is moving up in system architecture value chain."

Capacity Constraint and the CapEx Supercycle

The company is effectively capacity-constrained, with utilization in the 80-85% range across most lines. This has prompted an unprecedented capital spending plan. For 2026, ASE is raising total CapEx to about $10.5 billion (from an earlier $8.5 billion), split $4 billion for facilities and $6.5 billion for equipment. This includes 13 greenfield and 8 brownfield projects currently underway—a scale that would strain any organization. Joseph Tung admitted, "We have 13 greenfield projects going on. We have another 8 brownfield projects... it puts a lot of pressure on us in terms of execution." The investment is not just for LEAP; general packaging and test are also running tight, and the company is seeing strong demand in industrial, power, and connectivity devices. The market has taken notice. In the 30-day tape, AI-infrastructure-related names have wobbled (see the decliners list: High performance computing and co packaged optics are both in the red), but ASE's own guidance suggests the opposite—a sustained boom. This divergence may reflect the market's caution after a 40% run in ASE shares over the past year, but the call's concrete numbers are hard to dismiss.

LEAP: The New Growth Engine

The heart of the story is the rapid scaling of LEAP services, which include advanced packaging (os, full process, CoWoS-like) and testing. In 2025, LEAP brought in ~$1.6 billion; this year, it's tracking to over $3.5 billion, and next year it's targeted to double again. The incremental revenue is coming from both a broader customer base (not just the lead foundry's accelerator) and a higher mix of full-process services, which are more margin-accretive. Joseph noted, "We are on track with our full process business development... by next year, I think we will have pretty substantial growth in that area." The company is also investing in next-generation technologies like panel-level packaging (310x310mm line starting Q1 2027) and co-packaged optics, with Dr. Wu promising more details "in 2 quarters." This is a marked acceleration from prior calls. In October 2025, the tone was still cautious—"We'll give you a ballpark number saying that we will be having maybe at least TWD 1 billion revenue growth"—whereas today it's a hard commitment to double. The market should read this as a fundamental shift, not just cyclical strength.

A Pure-Play Advantage

ASE's key edge, repeated throughout the call, is its pure-play OSAT status—no conflicts with foundry or substrate vendors. As Dr. Wu explained, “We have no conflict with anybody. Therefore, we have a good way not only to collaborate with our customers long term, also collaborate with all of the ecosystem players that will turn critical for the overall complex integrated nature for the AI transition.” — Charlie Chan, Analyst · 2026-07-30 This positioning allows ASE to seize whatever packaging technology wins—CoWoS, EMIB, or panel—without betting the company on a single horse. The financials are beginning to reflect this strategic clarity. Q2 operating margin expanded 4.3 points to 11.1%, and the company expects Q4 ATM gross margin to exceed its previous 30% structural ceiling. The question investors will wrestle with is whether the massive CapEx and negative free cash flow (quarterly CapEx exceeds EBITDA) will deliver returns. Joseph was direct: "We will continue to have very heavy CapEx for this year, not only this year, but also going into next year. So I think the negative cash flow situation will remain for some time. But at the same time, we are still maintaining a very healthy balance sheet." In the prior call, management cautioned about multiple uncertainties, but the tone has shifted to confidence. In February 2025, Joseph noted, "We'll give you a ballpark number saying that we will be having maybe at least TWD 1 billion revenue growth." Now he is promising a doubling. This is a tell that the AI packaging opportunity is real and compounding.

The Bottom Line

ASE is no longer just an OSAT—it is the linchpin of AI hardware scaling. The company is betting billions that the AI-driven demand for complex packaging and testing will outlast the current cycle. With utilization high, pricing power returning, and a clear line of sight to 2027 LEAP revenue, the strategy looks well-founded. The risks are execution (20 construction projects at once) and the cyclicality of memory pricing (which is boosting EMS revenue but with lower margins). Yet if the LEAP doubling materializes, ASE's earnings power could be transformed. This is a company riding the AI wave with a specific, credible plan—and the market is listening.