AMD Pushes Outcome-Driven AI Infrastructure Strategy

AMD is targeting enterprise AI infrastructure as organizations shift from experimentation to production deployments focused on measurable business outcomes. Token economics are reshaping priorities, with companies increasingly mixing frontier and open-weight models based on workload needs rather than defaulting to costly frontier models for every task. AMD corporate vice president Suresh Andani emphasized that enterprises are paying for outcomes, not chips, driving demand for efficient on-premises infrastructure to run open-weight models securely and cost-effectively. AMD aims to deliver a full stack spanning GPUs, CPUs, networking, control planes and application layers to meet varying enterprise needs.