Intelligence + Continual Learning = Expertise — Yu Su, NeoCognition
AI agents excel at coding but fail broadly due to lacking expertise, not intelligence
“why we are so successful at the coding agents, but they're so terrible at anything else”
Yu Su (Ohio State / NeoCognition COO) argues that current LLM-based agents have human-like general intelligence but lack domain expertise, which explains why coding agents work well while agents in other domains fail. He frames token inefficiency and narrow agent success as symptoms of this intelligence-vs-expertise gap. The talk is conceptual groundwork for continual learning as the missing ingredient to build truly expert agents.