By JOHN BLYLER , Senior Contributing Editor, Semiconductor Digest For decades, semiconductor sustainability was treated largely as a chip design problem: better performance per watt, specialized accelerators and smarter power management. With the AI build-out driving a wave of new fabs, John Blyler argues the burden is now shifting onto manufacturing. Each new node demands more water, energy, materials and chemicals, and imec estimates that without sufficient intervention, IC manufacturing could account for as much as 3% of global greenhouse-gas emissions by 2040. Drawing on IDA Ireland, JST, ZEISS SMT and TechInsights, Blyler lays out the response: co-optimizing across design and the fab, shifting to renewable energy and better water recycling, adopting lower-emission process gases, and improving yield and equipment longevity so fewer wafers and less material are wasted. Efficiency gains won’t necessarily lower absolute emissions, though. TechInsights projects that between 2028 and 2030, AI accelerator shipments rise only 2.8% while modeled manufacturing emissions rise 40.4%, as the product mix shifts toward more carbon-intensive architectures. One promising route, he writes, is a feedback loop: turning AI itself, through process control, predictive maintenance, virtual fabs and digital twins, on the footprint of making the chips that AI demands. Click here to read the full article in the SEMICON West Show Daily. The post Has AI Shifted the Sustainability Balance from Chip Design to Manufacturing? appeared first on Semiconductor Digest .