Gary Marcus:成果来自人机数月反复协作,并非一句提示词完成
Gary Marcus 本人在 X 上澄清:大家想象中的一句提示词搞定的成果,实际是人和模型来回磨了好几个月。想了解 AI 产出背后真实工作量、别被宣传话术带偏的,可以看看这条短评。
Gary Marcus 在 X 上发文指出,外界讨论的这项成果并非像许多人想象的那样,靠给机器一句简单提示词就一步生成,而是经历了长达数个月的人机来回迭代。他强调 "months of back and forth between human machine",即人类与模型之间持续多轮的修改、反馈与验证过程。这条推文收获约 1400 次浏览、7 个点赞,讨论热度尚在早期阶段。
这条观点针对的是当前流行的一种误解:把 AI 产出的成果简化为单次提示词的产物。Marcus 的说法提醒读者,真实工作流往往包含人类多轮监督、纠错与整合,模型输出只是环节之一。对关注 AI 能力边界与人类贡献占比的人来说,这条短评提供了一个来自知名批评者的第一手观察角度。
also, seems like there was months of back and forth between human machine, not one simple prompt to the machine as many people are imagining.📊 3 ⚡
This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Gary Marcus, prompt engineering, human-AI collaboration.
Source attribution: curated from Gary Marcus (@GaryMarcus).