From AI to automotive, the industry has moved beyond debating chiplets to building at scale
Jul. 29, 2025 –
The chiplet era is underway. Walking through Moscone Center in San Francisco recently during the 62ndDAC, one thing jumped out at me: that reality was impossible to ignore – not just in the technology demonstrations, but in the conversations happening between foundries, IP vendors, OEMs and design houses. We’ve moved from asking “if” chiplets will scale to collaboratively solving “how” we design, validate and deploy multi-vendor systems across an integrated ecosystem.
That shift in tone – from hypothetical to hands-on – was just as clear on stage at the EE Times Future of Chiplets Summit at DAC, where I had the chance to give opening remarks and, later, join a panel of industry leaders. Across both sessions, I shared what I see as the three key areas we need to focus on if we want to move chiplets from bespoke engineering projects to scalable industry practice.
To unlock the full potential of chiplets, we must commit to three foundational imperatives:
Elegant hardware is meaningless without scalable software enablement. Specifically, software must remain stable across multiple IP generations. Developers must be able to program, provision and deploy on chiplet-based designs without complexity. That means defining common protocols, modular firmware and consistent software interfaces.
Today, most chiplet designs are internal to a single vendor. But real scalability comes when chiplets become products – reusable, interoperable and composable across different design points and even generations. That requires deeper integration between foundries, IP vendors, packaging partners and OEMs.
Interfaces like UCIe have made huge strides, but they only solve part of the puzzle. System-level specifications – how chiplets communicate, boot, share memory and expose telemetry – are where true interoperability lives. That’s the vision behind Arm’s Chiplet System Architecture (CSA), now supported by over 70 partners.