Do AI Tokenomics Matter More Than Model Benchmarks? with Chris Potts - #776

Stanford professor and Big Spin co-founder Chris Potts joins Sam Charrington to discuss AI tokenomics and his research into “tokenflation”—the possibility that token usage is growing faster than the measurable value those tokens produce. They explore how to measure return on AI spending, why benchmarks alone give an incomplete picture, and what inference-time scaling means for the economics of increasingly capable models. Chris also shares why expert AI users get better results by challenging and iterating with models, and how AI fluency affects outcomes. The conversation covers DSPy, interpretability, the limits of today’s transformer architectures, and where Chris sees opportunities for more fundamental innovation in AI.