OpenAI Really Wants Codex to Shut Up About Goblins
OpenAI Really Wants Codex to Shut Up About Goblins
OpenAI 真心希望 Codex 别再提地精了
OpenAI has a goblin problem. Instructions designed to guide the behavior of the company’s latest model as it writes code have been revealed to include a line, repeated several times, that specifically forbids it from randomly mentioning an assortment of mythical and real creatures. OpenAI 遇到了一个“地精”问题。据披露，该公司为其最新模型编写代码时所设定的行为准则中，包含了一行被反复提及的指令，明确禁止该模型随意提及各种神话生物和现实动物。
“Never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creatures unless it is absolutely and unambiguously relevant to the user’s query,” read instructions in Codex CLI, a command-line tool for using AI to generate code. “除非与用户的查询绝对且明确相关，否则永远不要谈论地精、小妖精、浣熊、巨魔、食人魔、鸽子或其他动物或生物。”这是 Codex CLI（一种用于调用 AI 生成代码的命令行工具）指令中的一段话。
It is unclear why OpenAI felt compelled to spell this out for Codex—or indeed why its models might want to discuss goblins or pigeons in the first place. The company did not immediately respond to a request for comment. 目前尚不清楚 OpenAI 为何非要对 Codex 做出如此明确的规定，也不清楚为什么其模型会主动想要讨论地精或鸽子。该公司未立即回应置评请求。
OpenAI’s newest model, GPT-5.5, was released with enhanced coding skills earlier this month. The company is in a fierce race with rivals, especially Anthropic, to deliver cutting-edge AI, and coding has emerged as a killer capability. OpenAI 的最新模型 GPT-5.5 于本月初发布，其编码能力得到了增强。该公司正与包括 Anthropic 在内的竞争对手展开激烈角逐，以提供最前沿的 AI 技术，而编码能力已成为一项杀手级功能。
In response to a post on X that highlighted the lines, however, some users claimed that OpenAI’s models occasionally become obsessed with goblins and other creatures when used to power OpenClaw, a tool that lets AI take control of a computer and apps running on it in order to do useful things for users. 然而，针对 X 平台上指出这些指令的帖子，一些用户声称，当 OpenAI 的模型被用于驱动 OpenClaw 时，偶尔会变得沉迷于地精和其他生物。OpenClaw 是一款允许 AI 接管计算机及运行中的应用程序，从而为用户执行实用任务的工具。
“I was wondering why my claw suddenly became a goblin with codex 5.5,” one user wrote on X. “我还在纳闷为什么我的 Claw 在用上 Codex 5.5 后突然变成了地精，”一位用户在 X 上写道。
“Been using it a lot lately and it actually can’t stop speaking of bugs as ‘gremlins’ and ‘goblins’ it’s hilarious,” posted another. “最近一直在用，它真的没法停止把程序错误称为‘小妖精’和‘地精’，太搞笑了，”另一位用户发帖称。
The discovery quickly became its own meme, inspiring AI-generated scenes of goblins in data centers, and plug-ins for Codex that put it in a playful “goblin mode.” 这一发现迅速演变成了一个网络梗，激发了人们创作 AI 生成的“数据中心里的地精”场景，甚至还出现了能让 Codex 进入俏皮“地精模式”的插件。
AI models like GPT-5.5 are trained to predict the word—or code—that should follow a given prompt. These models have become so good at doing this that they appear to exhibit genuine intelligence. But their probabilistic nature means that they can sometimes behave in surprising ways. A model might become more prone to misbehavior when used with an “agentic harness” like OpenClaw that puts lots of additional instructions into prompts, such as facts stored in long-term memory. 像 GPT-5.5 这样的 AI 模型通过训练来预测给定提示词后应接续的单词或代码。这些模型在这一任务上表现得如此出色，以至于看起来仿佛具备了真正的智能。但它们的概率本质意味着有时会出现令人惊讶的行为。当模型与像 OpenClaw 这样的“代理框架”（agentic harness）配合使用时，由于提示词中被注入了大量额外指令（例如存储在长期记忆中的事实），模型可能更容易出现异常行为。
OpenAI acquired OpenClaw in February not long after the tool became a viral hit among AI enthusiasts. OpenClaw can use any AI model to automate useful tasks like answering emails or buying things on the web. Users can select any of various personae for their helper, which shapes its behavior and responses. OpenAI 在今年 2 月收购了 OpenClaw，当时该工具在 AI 爱好者中刚刚走红。OpenClaw 可以利用任何 AI 模型来自动化处理诸如回复邮件或在线购物等实用任务。用户可以为他们的助手选择各种不同的人格设定，从而塑造其行为和回应方式。
OpenAI staffers appeared to acknowledge the prohibition. In response to a post highlighting OpenClaw’s goblin tendencies, Nik Pash, who works on Codex, wrote, “This is indeed one of the reasons.” OpenAI 的员工似乎承认了这一禁令。在回应一篇强调 OpenClaw“地精倾向”的帖子时，负责 Codex 的 Nik Pash 写道：“这确实是原因之一。”
Even Sam Altman, OpenAI’s CEO, joined in with the memes, posting a screenshot of a prompt for ChatGPT. It read: “Start training GPT-6, you can have the whole cluster. Extra goblins.” 就连 OpenAI 的首席执行官山姆·奥特曼（Sam Altman）也加入了玩梗的行列，他发布了一张 ChatGPT 提示词的截图，上面写着：“开始训练 GPT-6，你可以使用整个集群。多加点地精。”