This episode will be available to stream at 11am ET.

The rapid growth of AI is transforming how the industry thinks about compute as the benefits of traditional chip scaling begin to slow. This session will explore why GPUs have become so dominant, what new accelerator architectures are emerging, and where the next major gains in AI and HPC performance could come from. The discussion will examine the growing specialization of compute, from silicon and advanced packaging to memory, software and energy efficiency, as well as how organizations are navigating an increasingly diverse and fast-changing accelerator landscape. This episode will uncover:

• Whether GPUs will continue to dominate AI training and high-performance computing
• How the slowing of Moore’s Law is driving greater specialization in chips and accelerators
• How new architectures are being matched to different AI and HPC workloads
• How software, memory, packaging and energy efficiency are shaping the next generation of compute