With the AI boom showing no signs of going away any time soon, the hardware running new data centers continues to evolve at a rapid pace.

GPUs have been the name of the game for many developers during the first stage of the AI data center build-out, but as the market evolves, and AI companies bank on greater adoption by enterprise, other chips are becoming increasingly important in the stack.

Read this supplement for a deep dive on chip production, AI at the network Edge, and cooling strategies for AI hardware, featuring articles on:

• LPUs, CPUs, and life beyond the GPU: An interview with Nvidia's Vice President of Hyperscale and High-Performance Computing Ian Buck about Groq and agentic AI
• Liquid cooling for AI data centers: Implementing it successfully to meet the power and GPU-intensive demands of high-density data centers
• From design to manufacturing: Why Arm is making its own chip for companies without the capital required to develop their own processors
• Bringing Edge AI hardware to the world: Axelera AI's founder discusses his vision