• Three-pillar AI strategy: AMD is driving its AI roadmap across CPUs, GPUs, and adaptive SoCs by releasing yearly leadership products, fostering open-source software partnerships, and co-designing systems with hyperscale leaders like Meta and OpenAI.
• Inference efficiency gains: With monetization shifting toward inference, AMD targets roughly 10x generation-over-generation improvements in tokens produced per dollar and per megawatt on a one-year product release cadence.
• Agentic compute demand: Autonomous multi-agent coordination requires rapid orchestration and task execution, driving renewed structural demand for CPUs to manage workloads alongside GPUs.
• India investment footprint: Backed by a $400 million commitment through 2028, AMD employs over 10,000 personnel in India who contribute directly to chip design, tape-outs, and open-source software across the entire product portfolio.