AI’s growth is straining semiconductor and data center materials, pushing demand for advanced chemistries

The rapid expansion of AI workloads is pushing semiconductor and data center hardware toward fundamental physical limits in performance, thermal management, electrical efficiency, and reliability, according to Mike Finelli, chief technology and innovation officer and chief North America officer at Syensqo. These constraints are elevating the importance of advanced materials that can meet stricter requirements for high temperature tolerance, chemical and plasma resistance, purity, and long-term stability.

Finelli describes a hierarchy in which materials sit at the “top of the pyramid,” determining what becomes feasible for AI systems. He argues that advanced materials are no longer just enablers but active constraints—or accelerators—of AI capability. Syensqo is developing materials for high-voltage data center architectures, advanced sealing solutions for semiconductor manufacturing, and thermal-management fluids for direct immersion cooling.

The company also sees cross-industry spillovers, such as materials originally designed for electric vehicles being adapted to address rising voltage and energy-density demands in data centers. This transfer of innovation highlights how materials advances in one sector can unlock performance gains in another.

Sustainability is being redefined in materials R&D. Finelli says Syensqo aims to eliminate the traditional trade-off between performance and environmental impact by embedding sustainability goals into the earliest stages of research, rather than retrofitting them later.

AI is also transforming how new materials are discovered. Syensqo uses AI agents to digitally synthesize millions of molecular combinations, predict their performance and sustainability profiles, and winnow candidates down to a smaller set for lab validation. This approach allows researchers to explore a broader design space more quickly and focus human expertise on complex engineering challenges.

Finelli envisions a reinforcing cycle: AI-driven materials discovery improves AI infrastructure, which in turn enables more sophisticated AI systems that can accelerate materials science. He calls this an “accelerated materials innovation cycle” that could expand the frontier of what future technologies can achieve.