AI Agents Speed Up Research but Complicate Evaluation
Westlake University researcher Wang Huan said at the 1024 Model Force community forum that AI agents are accelerating literature review, coding, experiments, paper writing and review. He also argued that faster output does not necessarily mean faster scientific progress, because claims still require verification and peer review, placing greater pressure on research evaluation systems.
InfoQ Chinese's report of Wang's presentation highlights a central challenge in AI-assisted research: productivity can rise while assessing reliability, novelty and genuine contribution becomes harder.

在1024 模力工场 · AI社区日"主论坛上,西湖大学工学院助理教授、博士生导师、高效智能计算实验室(ENCODE Lab)负责人王欢,以《AI智能体对科研的影响》为题作主题分享。

王欢有两个重点研究方向:Efficient AI(高效人工智能),关注如何提高 AI 的运行效率、降低资源消耗;AI Evaluation(AI 评测),关注如何定义和衡量智能。
这次分享中,他结合计算机和 AI 研究的经历,讨论了 Agent 带来的两方面影响。一方面正好也关于“效率”:读文献、写代码、做实验等工作正在加速;另一方面则关于“评价”:论文更容易产出,也给成果鉴别、同行评审和科研评价带来了更大压力。