While GPUs dominate AI hardware discussions, the proliferation of AI agents is causing a significant, often overlooked, CPU shortage. Agents rely on CPUs for web queries, data processing, and other tasks needed to feed GPUs, straining existing infrastructure and driving new demand for companies like Arm and Intel.

The shift from simple chatbots (one user request, one API call) to agentic AI systems will decouple inference requests from direct user actions. A single user request could trigger hundreds or thousands of automated model calls, leading to an exponential increase in compute demand and cost.

Ben Thompson argues the shift from simple chatbots to AI agents creates an exponential, non-speculative demand for compute. Agents automate complex, multi-step tasks, driving constant usage that justifies the massive capex investments by hyperscalers. This suggests the current spending is based on real demand, not bubble-fueled speculation.

The current AI data center arms race isn't about meeting today's demand for chatbots. It's fueled by companies like Meta betting on a future where personal AI agents run constantly, analyzing every interaction. This vision of persistent, parallel agents requires an exponential increase in compute, explaining why they will buy any available capacity.

The tangible utility of agentic tools like Claude Code has reversed the "AI bubble" fear for many experts. They now believe we are "underbuilt" for the necessary compute. This shift is because agents, unlike simple chatbots, are designed for continuous, long-term tasks, creating a massive, sustained demand for inference that current infrastructure can't support.

Jensen Huang quantifies the massive computational leap required for advanced AI. The move from generative AI to reasoning was a 100x compute increase, and the subsequent move to agentic systems that can perform work represents another 100x jump. This results in a staggering 10,000x increase in computational demand in just two years.

The largest driver of future energy consumption for AI won't be human-initiated queries on chatbots. Instead, it will be the massive, continuous "machine-to-machine" traffic generated by autonomous AI agents performing tasks, which will ultimately swamp human-AI interaction and create a runaway demand for compute power.

After the current memory crunch, the next AI infrastructure bottleneck will be CPU and networking. The complex orchestration required for emerging agentic AI systems will strain these resources, a trend already visible in companies like Fastly seeing demand spikes just for workload orchestration.

While GPUs get the headlines, AI expert Tae Kim warns of a major coming CPU shortage. The complex orchestration, tool calls, and database queries required by AI agents are creating huge demand for CPU cores, a trend confirmed by major chipmakers and hyperscalers.