Applied Materials opens $5B EPIC chip research facility
Two stories from the hardware layer of AI today: the biggest chip-equipment maker just opened a shared research megafacility in Silicon Valley, and an industrial automation incumbent is buying its way into physical AI.
Applied Materials opened the EPIC Center, a $5 billion shared chip research facility in Silicon Valley, with TSMC, Intel, Samsung, Micron and SK hynix signed up as founding partners. The 600,000-square-foot site in Sunnyvale carries nearly four acres of cleanroom space where engineers from rival chipmakers will work side by side on 3D architectures, advanced packaging, low-resistance wiring and hybrid bonding — the process technologies that determine whether the next generation of AI silicon ships on time. CEO Gary Dickerson told the New York Post the goal is to cut the development cycle by at least 30%, a meaningful number because traditional chip R&D can run up to a decade and AI compute demand is not waiting; the company's own release puts the same target more softly, as compressing the timeline from materials research to high-volume manufacturing. Intel CEO Lip-Bu Tan framed the stakes at the opening: "We see a path to more than a trillion transistors in a package by 2030," he said, arguing that rising R&D costs force equipment suppliers and chipmakers to work closer together. The takeaway: at atomic scale, the laboratory itself is becoming competitive infrastructure — and the real headline is the partner roster, not the construction budget. Applied Materials says the $5 billion capital investment is expected to scale over time as customer projects ramp, so it is a target, not money already spent.
Omron is buying US industrial PC maker OnLogic and its parent Interlogic for $675 million, a deal the Japanese automation giant says will power its physical AI push. Reuters reported the agreement, with Nikkei Asia adding that the acquisition could close as early as January; OnLogic keeps its brand and management while Interlogic becomes a wholly owned subsidiary. The logic is straightforward — Omron's precision control technology plus OnLogic's edge computers is the stack needed to sell "physical AI" platforms to factories and to expand its US customer base beyond its current industrial base. It is another incumbent paying real money for the edge-computing layer that robots and autonomous systems run on, and a signal that physical AI M&A is consolidating around hardware access rather than model talent.
What to watch: which EPIC partner announces the first process breakthrough out of the building, and whether the Omron deal clears regulators on the January timeline.
Do chipmakers gain more from shared research floors than from racing each other in isolation? Tell us in the comments.
