Handing the baton at imec: same compass, bigger engine

As of April 1, 2026, longtime CEO Luc Van den hove becomes Chairman; Patrick Vandenameele steps in as CEO. In Leuven, the duo outlined imec’s next decade: expand the world’s leading R&D pilot line, double down on AI-driven compute efficiency, and turn Europe’s strengths into sovereignty through design, not just fabs.

After 17 years at the helm, Luc Van den hove will move upstairs to become Chairman of imec’s Board of Directors. His successor, insider-entrepreneur Patrick Vandenameele, won’t be rewriting the playbook. “Don’t expect me to come in here in April and change everything,” Vandenameele said. “We’ve been working together on this strategy for the last few years, and I think we have a more relevant strategy than ever.”

The three assets imec won’t compromise on

Van den hove opened his clarifications, last week in Leuven, with first principles. Imec exists, he said, to deliver “disruptive innovation” in chip technology and the applications it enables, because almost every deep-tech challenge “builds on progress in chip technology.” The institute’s edge is a compound of three assets: top-notch infrastructure, top talent, and a uniquely dense ecosystem of partners spanning materials, equipment, foundries, fabless design, systems companies, and universities.

That formula has produced a business model unusual for public research: more than three-quarters of revenue now comes directly from industry, with government support acting as the long-term anchor. “I personally don’t know any other research organization in our field that has such a strong industry endorsement,” Van den hove noted.

Capacity is strategy: Fab4 in Leuven, Fab5 in Spain

If imec’s pilot line is the institute’s beating heart, the coming years are about giving it more muscle. With major backing under the European Chips Act, imec will add €2.5 billion in tools and 6,000 m² of cleanroom capacity in Leuven. Fab4, physically linked and automated with the current Fab3, will be equipped with “the full suite of next-generation ASML tools like the High-NA EUV machine.” Construction is set to start next year, with the new facilities targeted to be ready by the end of 2028.

A second strand of the expansion - Fab5 in Spain - will introduce specialty devices onto 300-mm wafers for application-rich domains, such as healthcare and AR/VR. Timing mirrors Fab4: construction to start next year, with readiness early 2029. Why the extra capacity? Advanced processes are more complex and take longer to run; volume, cycle-time, and operational excellence become competitive levers. “We are committed to making sure that this infrastructure remains the most advanced R&D pilot line on the planet,” Van den hove said.

AI’s energy wall and the scaling roadmap

Much of imec’s roadmap is being pulled by AI’s voracious demand for compute. Left unchecked, that demand translates into unsustainable energy. The answer, in imec’s view, is relentless efficiency: extend 2D scaling, push new transistor architectures, and move into the third dimension.

“Many people have been saying Moore’s law is dead,” Van den hove said. “There is this phenomenal demand for more advanced technologies… and that creates this enormous push to extend Moore’s law.” In addition to lithographic scaling, including the High-NA EUV transition, imec is driving 3D approaches: stacking NMOS and PMOS devices, stacking chiplets (memory on logic), and interconnecting at scale via advanced packaging. Critical to that third dimension is photonics for chip-to-chip links, moving from pluggable optics to chip-package optics and photonic interposers that act like waveguide highways between chiplets.

Imec dubs this blended future “CMOS 2.0”: 2D + 3D + optical interconnects, enabling “compute systems with close to a trillion transistors” without blowing the power budget.

From cars to clinics: where chips become systems

Two application arenas headline imec’s next push:

• Automotive. As vehicles electrify and edge toward autonomy, sensors multiply and in-car data explodes. The sector needs smaller, cheaper cameras, radars, lidars, and a high-performance compute platform that’s cost-effective across model variants. Imec’s answer is reference designs, standardization and quality work, and a new design hub in Heilbronn, Germany to pull the supply chain closer and accelerate innovation. “We really need to bring together all of the key players… and bring the car manufacturers much closer to the chip innovation process,” Van den hove said.
• Health & bio. By converging electronics, photonics, sensors, and microfluidics, imec is shrinking lab-grade functions onto chips: from DNA sequencing (aiming for 100×–1000× cost reductions) to neuro-probes and organ-on-chip platforms such as blood–brain barrier models that can generate reliable, human-relevant data for AI-guided drug discovery. “We’re just at the start of this revolution,” Van den hove said.

Fueling Europe’s sovereignty by designing here

Asked about digital sovereignty, Van den hove was clear-eyed: collaboration across borders is non-negotiable, yet Europe must grow its value-chain share by building on strengths. “You cannot make any chip in the world without ASML technology and imec processes,” he said. The next step, in his view, is design capability on leading-edge nodes. Focusing only on advanced fabs misses the point if the chips used worldwide continue to be designed elsewhere. Imec’s answer is to incubate and enable more European fabless players through joint R&D programs, IC-Link prototyping and design services, and venturing.

That venturing engine has picked up speed. One recent example is Vertical Compute, a startup tackling AI’s memory-compute bottleneck by “bringing more memory very close to logic processors,” launched with €20 million in initial capital. The broader pipeline is sizable: hundreds of startups supported annually through IC-Link, and dozens created or backed via imec’s venture programs.

Culture as an operating system

For all the hardware, imec insists its culture is the ultimate differentiator: integrity (essential when competitors co-develop under one roof), connectedness (linking companies, universities, and disciplines from nano- to biotech), passion, and excellence. “If you can’t be among the best in the world in a certain subject, we'd better change our strategy,” Van den hove said.

In April, Van den hove's successor will take over the baton. Vandenameele’s biography reads like imec’s full stack: PhD in Leuven, four startups (three built on imec tech), exits into U.S. and China corporates, a return to build imec’s venturing practice and the imec.xpand fund, then a move into corporate strategy and co-COO. That 360-degree view, from lab to fab to market, is the point. “It’s exceptional how we’ve become the trusted partner of the entire industry,” he said. The job now is multiplication: keep scaling the basic transistor platform while deepening ties with hyperscalers and system companies that increasingly set device roadmaps.

Patrick Vandenameele, imec

The handover is designed for continuity: a six-month transition to April 1, 2026, with Van den hove staying close as Chairman focused on external and stakeholder management. “A gradual smooth transition focusing very much on continuity and stability is very important in order to keep this trust relationship with all our partners,” he said. “We believe Patrick is the right person with the right skills, the right background… and the right values.”

In other words: same compass, bigger engine. With Fab4 and Fab5 coming online at the turn of the decade, a compute roadmap that blends 2D, 3D and photonics, and a venture flywheel seeding Europe’s fabless future, imec is betting that the shortest path to digital sovereignty runs straight through Leuven’s pilot line, and the ideas it puts into the world.