Tokenomics: The AI cost challenge telcos can't afford to ignore

Join AT&T's Mark Austin and TM Forum's Guy Lupo for a discussion on one of the telecom industry's most pressing AI challenges: tokenomics. As AI adoption accelerates, operators are seeing explosive growth in token consumption, making cost management, governance and observability critical business priorities. Drawing on AT&T's real-world experience, Mark shares how the company is tackling AI costs at scale through intelligent routing, FinOps practices and the TM Forum Model as a Service (MODaaS) framework. The conversation explores why AI gateways must evolve beyond simple routing, how operators can optimize model selection without compromising performance, and what practical lessons have emerged from large-scale deployment. They also discuss the growing importance of specialized telco AI models, sovereign AI, and the role of trust in enabling enterprise-wide AI adoption. Together, they outline how operators can build the foundations for the autonomous, AI-native enterprise and compete successfully in the Race to 2030.