As computing demands increase, particularly with the rise of artificial intelligence (AI), understanding which resulting hardware waste poses the greatest threats to human and ecological health is essential for sustainable design. Currently, computer system designers lack quantitative tools to assess the toxicity impact of their design choices. Through an interdisciplinary collaboration, the research team will develop SysTox, a novel data-driven framework that quantifies the toxicity impact of system design choices. This approach serves the national interest by advancing computing technology, reducing harmful e-waste, and providing a foundation for responsible technological advancement. The project will develop methodologies to analyze computing systems at the component level, identifying the elemental composition of various components and assessing their impacts in the waste stream. The research team will create new datasets through experimental measurements, develop models to quantify toxicity impacts, and establish an optimization framework that can guide computer hardware designers in making less toxic design choices while maintaining performance. The resulting framework will enable computer system engineers to make quantitative tradeoffs between performance, efficiency, and footprint, transforming how architectures and systems are designed across scales from microprocessors to data centers. This award reflects NSF's statutory mission and has been deemed worthy of