Intel said the balance of AI computing is shifting, with CPU and GPU deployments now near parity on a unit basis as workloads move from training to inference. CFO David Zinsner cited second-quarter data-center AI revenue of $6.3 billion, up 59% year over year, and record server growth driven by the fast-ramping Xeon 6 line, with demand outstripping supply. CEO Lip-Bu Tan said general-purpose server CPU density is rising as AI expands to agentic and multi-agent systems. While the company stopped short of endorsing a specific forecast that puts the CPU market at $220 billion by 2030, the remarks underscore a potential rebalancing of AI spending—good news for Intel’s core franchise and a signal to Nvidia investors that inference at scale could diversify infrastructure beyond GPUs even as training demand remains strong.
Related article:
