I was curious how X is fighting AI slop, so I looked at its open-source algorithm. It has a behavioral model called TweetSpamBot that analyzes up to 512 recent account actions, looking at signals like posting bursts,…

X 在 8 月 13 日开源 For You 时间线排序代码后,开发者发现其反 AI 垃圾内容主要依赖一个名为 TweetSpamBot 的行为评分模型,能识别高频发帖、批量引用转推等模式,但几乎不分析帖子文本,导致刻意包装的 AI 内容农场仍能绕过限制。
![The evolution of a workday: Before AI: [decision] [process] [process] [decision] [process] [process] [process] 10pm: still going With AI: [decision] [decision] [decision] [decision] [decision] 3pm: [brain empty] [need...](https://www.chat-gpts.plus/wp-content/uploads/2026/08/ai_cover_1-512-768x403.jpg)
![[Bug]: Using the official interface to call the agent application in Ragflow, the parameters in the initial component cannot be passed in](https://www.chat-gpts.plus/wp-content/uploads/2026/08/9617-f9a4f153-768x403.jpg)
![[Question]:GPT-OSS with JSON Structured Output caused error for LLM](https://www.chat-gpts.plus/wp-content/uploads/2026/08/9462-601c3b24-768x403.jpg)
![[Bug]: If you click the pause button before the AI finishes its response, it will stop replying to you.](https://www.chat-gpts.plus/wp-content/uploads/2026/08/9641-3017b443-768x403.jpg)




