论文：换代模型发布后，AI文本检测器识别率从99%跌至3.8%
精选理由
检测器换代前抓99%的AI改写文本，换代后只剩3.8%，用AI检测工具的人真该看看这个数据。
一项研究发现，基于厂商旧模型训练的AI文本检测器，在新一代模型发布前能识别超过99%的改写文本，发布后识别率仅剩3.8%。这意味着用于筛查科研论文的AI写作检测工具会随LLM换代而失效。作者建议每次新模型发布后都要重新测试检测器的实际效果。
原文 · rohanpaul_ai
This paper finds that AI-text detectors trained on a vendor's older models caught over 99% of rewrites before a generation change and only 3.8% after it.
Detectors that screen scientific papers for AI writing can stop working when a new LLM generation arrives, so anyone relying on them should re-test them with every model release.
