NVIDIA told us exactly where AI is going — and almost everyone heard it wrong

CES 2026 signals a shift towards optimizing the supply chain for always-on AI, marking the beginning of an industrial cycle focused on delivering AI cheaply and reliably at scale. NVIDIA's launch of the Rubin platform at CES, a rack-scale system designed around token economics, slashes inference token generation costs while efficiently serving large models. OpenAI's strategic moves, including partnerships with NVIDIA, AMD, and Broadcom, along with cloud contracts with AWS and CoreWeave, guarantee its competitive edge in securing compute capacity. The scarcity created by OpenAI's deals is reflected in rising DRAM prices, underscoring the demand-constrained nature of the AI economy, where physical AI and ambient intelligence are set to increase inference demand and diversify AI applications.