About this book:

"Compute Rush: Inside the Global Race for AI Chips, Data Centers, and the Power to Run Them" offers an unparalleled deep dive into the physical infrastructure underpinning the artificial intelligence revolution. As AI models explode in complexity and capability, this insightful book reveals how the insatiable demand for "compute"—the raw power to train and run these models—is hitting the physical limits of our grids, factories, and cities. Explore the intricate world of silicon chips (GPUs and their challengers), the colossal steel structures of data centers, and the vast amounts of steam (power, water, and cooling) required to keep them running. Through a mix of expert interviews, case studies, and incisive analysis, "Compute Rush" uncovers the critical bottlenecks and strategic decisions shaping the future of AI.

The book meticulously dissects the core drivers of AI’s compute hunger, from the economics of training versus inference to the relentless pursuit of lower latency and the emerging data bottleneck. It unravels the complex global supply chain, revealing how a handful of semiconductor foundries control access to cutting-edge chips and how geopolitical forces, export controls, and industrial policies are reshaping the technological landscape. Crucially, "Compute Rush" addresses the colossal environmental and societal implications of this growth, examining the unprecedented demands on electricity grids, the critical role of cooling technologies, the competition for water resources, and the growing community pushback against massive data center campuses. This essential read illuminates why the future of AI won't just be written in code, but built—one chip, one data center, one megawatt at a time.

What You'll Find Inside:

• **The Compute Bottleneck:** Explore how the insatiable demand for AI chips (silicon), data centers (steel), and electrical power (steam) has created critical physical bottlenecks, redefining the limits of AI growth.
• **Training vs. Inference Economics:** Understand the distinct computational and economic demands of training new AI models versus running them for real-world applications, and how optimizing inference is key to long-term profitability.
• **Geopolitical Race for Chips:** Delve into the intense global competition for advanced semiconductors, including the dominance of foundries like TSMC, the critical role of export controls, and national industrial policies aimed at securing chip supply.
• **Energy Grid Under Strain:** Discover how AI's unprecedented and rapidly growing electricity consumption is overwhelming existing power grids, necessitating massive investments in generation, transmission, and the potential resurgence of nuclear power.
• **The Rise of Software Efficiency & Edge AI:** Learn how innovations in software optimization (quantization, sparsity, distillation) and the shift towards on-device and edge AI are crucial for reducing the cost of intelligence, improving privacy, and enabling sustainable, scalable AI deployment.

Who's It For:

This book is essential for anyone seeking to understand the foundational, physical realities driving the global AI boom. It targets technology executives, investors, policymakers, infrastructure developers, and energy professionals who need to navigate the complex interplay of hardware, energy, and geopolitics shaping the future of artificial intelligence. Readers interested in the supply chain, economic implications, and environmental impact of AI will find deep insights and actionable frameworks.