David Silver

David Silver appears in E226｜聊聊DeepMind创始人哈萨比斯：一个科学家与失控的AI竞赛 as an early collaborator of Demis Hassabis in games and later as a key figure in DeepMind’s reinforcement-learning route. The source says Silver aligned with Richard Sutton’s reinforcement-learning theory and studied at the University of Alberta.

In the episode’s narrative, Silver helps connect game intelligence to AlphaGo and the broader Reinforcement Learning AGI Path. That makes him part of the contrast with Language Model Scaling Bet: DeepMind’s early conviction was that agents learning from environments and rewards could generalize toward AGI, while language models later shifted the competitive center.

Connections

• Demis Hassabis, DeepMind, and Google DeepMind — company and collaborator context.
• Richard Sutton and Reinforcement Learning AGI Path — technical lineage emphasized by the source.
• AlphaGo — public proof point for DeepMind’s game-centered route.
• Language Model Scaling Bet — route contrast after ChatGPT.