经济学人：一个“幽灵写手”正在重塑世界

作者：经济学人      来源：一天一篇外刊

A GHOST WRITER is haunting the English language. The linguistic spectre can turn its hand to prose, poetry, journalese and corporate jargon. It is frightfully versatile: you can get it to mimic Shakespeare’s sonnets or a schlocky beach read; Ernest Hemingway’s taut prose or the office-printer manual. It is frightfully fast, churning out thousands of words a minute. (Hemingway rarely produced as many in a day, and required much more booze.) Wordsmiths are spooked.

“幽灵写手” 正游荡在英语世界。这位语言幽灵得心应手，散文、诗歌、新闻文体、企业话术无一不通。它本领惊人，既能仿写莎士比亚十四行诗，也能炮制粗制滥造的通俗读物；海明威凝练的文风，办公打印机说明书，同样信手拈来。它下笔速度更是骇人，每分钟便可产出数千文字。（海明威整日写都未必写得出这么多，还少不了大量酒水助兴。）文字从业者无不心生不安。

AI writing is everywhere. It is in your inbox and on your LinkedIn feed. It is all over the internet, drafting more than a third of new websites by one count. Large language models (LLMs) are helping students write essays and probably helping scientists write papers. Some allege AI-generated prose won the Commonwealth Short Story prize this year, with judges praising its “quiet authority”. (The Commonwealth Foundation denied the claim.)

AI写作无处不在。它藏你的收件箱，出现在领英动态之中。互联网上更是随处可见。有统计估算，超三分之一新建网站的内容由AI起草。大语言模型（LLM）既协助学生撰写文章，恐怕也在助力科研人员刊发论文。有人称，本年度英联邦短篇小说奖获奖作品为AI创作，但评委还盛赞文字 “从容厚重，自带说服力”。（英联邦基金会对此予以否认。）

LLMs have stylistic quirks. They are thought to maximise the use of long em-dashes—and the use of words like “maximise”. They like to “deep dive” (and, better yet, “delve”) into the “rich tapestry” of the world. AI writing is not about a single word or phrase, but a rich tapestry of things.

大语言模型存在独特的行文偏好。观察发现，大模型偏爱长破折号，也频繁使用 “最大化”这类词汇。它们喜欢对世间万象“深度探究”，更偏爱用“深挖” 一词，它们还常将世界比作一幅 “斑斓织锦”。AI写作的特质，并不体现在零星词句上，而是层层交织、如同织锦一般的语言整体。

Spotting AI texts can be tricky. This is in part because you need evidence beyond a few words or dashes: claiming that a text is by an LLM because it uses the word “delve” is like claiming one is by Jane Austen because it uses “imprudence”. Bots also write in slightly different ways. There is no single style of AI writing, explains Karolina Rudnicka, a linguist at the University of Gdansk in Poland, just as there is no single style of human writing. Writers have idiosyncrasies—Emily Dickinson, for instance, loved em-dashes—and bots may do, too.

辨别AI文并非易事。首先，仅凭个别词汇或是破折号不足以判定文本来源：若只因文中出现 “深入探究” 就断定文章由大语言模型生成，无异于看到“轻率”一词，便认定作品出自简・奥斯汀。不同 AI 程序的行文方式也存在差异。波兰格但斯克大学语言学家卡罗琳娜・鲁德尼茨卡解释，人类写作不存在统一风格，AI 写作亦是如此。创作者皆带有个人行文特质。艾米莉・狄金森是出了名偏爱长破折号，AI模型同样拥有身写作特征。

But there are a few ways to identify LLM-generated text. One is to use detection algorithms that are trained to spot the texture of human or AI prose. Pangram, a leading firm, claims to have 99.98% accuracy. (It has partnered with Substack, a blogging platform, on such a tool.) Detectors, however, are black-box algorithms that can give false positives. They do not give reasons for why they reach their conclusions.

不过，仍有若干手段识别大模型生成文本。第一种方式是依靠专门的检测算法。这类算法经过训练，能够识别人类文字与AI文本语言质感的区别。头部AI检测企业 Pangram 宣称检测准确率高达99.98%。（该企业已和博客平台 Substack 合作开发相关工具。）但检测器属于黑箱算法，有可能产生误报，并且不会说明判定依据。

作者：经济学人      来源：一天一篇外刊