2025-09-28 · Industry Analysis
Richard Sutton, widely regarded as the father of reinforcement learning (RL), critiques the dominance of large language models (LLMs), arguing they represent a technological dead end. Unlike RL, which learns from goals and real-world consequences, LLMs only mimic human language patterns without grounding in reality. Sutton envisions a future where intelligence develops through continual experiential learning, not predictive text. He situates AI as both humanity's successor and the next cosmic stage in the evolution of intelligence.