Episodebeskrivelse
Falling per-token costs aren't reducing AI infrastructure spend — they're accelerating it. In this episode, we break down why Jevons Paradox is playing out in real time across the AI supply chain, and what the input signals — from DRAM pricing to transformer lead times — are telling us about where this goes next.
We cover: why agentic AI workloads consume 5-30x more tokens per task than chatbots, the seven upstream signals all moving in the same direction, and three companies positioned across the full stack from silicon to switchgear — Marvell Technology (MRVL), Micron Technology (MU), and…
Nettsted: podcasters.spotify.com

