Sam Altman 关注 Goodfire 的模型可解释性研究,学界对其结论存争议
Sam Altman 盯上了 Goodfire 这家做模型可解释性的公司,想搞清楚模型内部到底在想什么。不过好几位知名科学家不买账,觉得结论站不住脚。做模型安全或对齐方向的人可以看看争议在哪。
据 The Information 报道,OpenAI 的 Sam Altman 对 Goodfire 的研究表现出兴趣,这项研究旨在帮助解释 AI 模型的内部运作方式。Goodfire 是一家专注模型可解释性(interpretability)的公司,其方法试图打开模型"黑箱",弄清模型生成某个输出时内部激活了哪些特征。
不过,该报道同时指出,多位 leading scientists 对 Goodfire 研究结论的有效性持质疑态度,认为其解释方法可能并不如实反映模型的真实计算过程。可解释性研究近年来成为安全与对齐讨论的重点方向,Anthropic、DeepMind 等机构也在投入类似工作,Altman 的关注显示 OpenAI 对这一路线的评估兴趣。
对于关注模型安全与对齐的从业者和研究者来说,这条动态提示了可解释性方向正从纯学术议题走向工业界重点关注的领域,但学界争议也说明该领域的方法论尚未收敛。
OpenAI’s Sam Altman is interested in research from Goodfire that could help explain how AI models think, even as leading scientists dispute its conclusions.
Read more: https://t.co/t1YMNQCF02
This report is categorized under AITOP's "Industry" 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 OpenAI, Goodfire, Sam Altman, interpretability.
Source attribution: curated from The Information (@theinformation). Cross-source corroboration: 11 independent sources have covered this development (including PolymarketMoney, kimmonismus, The Rundown AI), confirming industry-wide signal strength.