语言人工智能指南

发现人工智能的虚假引文

Chatbots invent citations because they generate text that looks right, not text they have checked.

  • 4 分钟阅读
  • 最后更新
在本页4 分钟阅读
  1. 概述
  2. 深入探讨
  3. 战略影响
  4. The Future of Spotting Fake Citations from AI
  5. 现实世界的实施
  6. 风险与防护栏
  7. 实施路线图
  8. 不断探索
  9. 常见问题

概述

A made-up reference can have real-sounding authors, a plausible journal and even a DOI. That matters because a fake or misattributed source in your paper can cost you the grade, damage your credibility, or in professional settings lead to formal sanctions.

深入探讨

A language model writes by predicting which words are likely to come next. Citations follow a very predictable pattern: author names, a year, a title, a journal, a volume, page numbers and often a DOI. A model has seen millions of real references, so it can produce one that looks perfectly formed and points to nothing. This is not the model lying on purpose. Unless it is connected to a search tool, it has no built-in step that checks whether the reference exists. Bad AI citations come in four kinds. Some are completely made up. Some mix real researchers with a title they never wrote. Some name a real paper but get details wrong, such as the year, the journal or the page numbers. The hardest kind to spot is a real, correctly cited paper that doesn't say what the chatbot claims it says. The consequences are real. In the 2023 case Mata v. Avianca in federal court in New York, lawyers filed a brief citing court cases that ChatGPT had invented, and the judge sanctioned them. To check sources before you submit, first search the exact title in quotes in Google Scholar or your library's database. Second, if there is a DOI, open it through doi.org and confirm the paper it leads to has the same title and authors. Third, check that the authors, year and journal all match. Fourth, open the source and find the passage that supports your claim. Fifth, if you can't verify a source, remove it. A common mistake is asking the chatbot to confirm its own citation. It can produce a confident 'yes' just as easily as it produced the fake reference. Tools that search the web lower the risk but don't remove it, because they can still misstate what a real source says.

战略影响

速度与规模

语言工作流程可以在不牺牲一致性的情况下更快地移动。

交通与覆盖范围

它扩展了跨语言和沟通方式的访问。

更清晰的判决

团队可以花更多时间进行判断,而自动化则可以处理重复。

The Future of Spotting Fake Citations from AI

More chatbots now search the web by default and show links, which reduces fully made-up references in everyday use. Misattribution remains a hard problem, though, because it takes careful reading to judge whether a source really supports a specific claim. Publishers, libraries and citation tools are building better ways to check references automatically, and some writing tools may flag unverifiable citations before you submit. None of this replaces the basic habit: find the source, open it and read the relevant passage yourself. Schools will likely keep holding students responsible for every citation, whatever tool suggested it.

现实世界的实施

A student asks a chatbot for five studies on screen time and sleep. When she searches Google Scholar for the exact titles in quotes, two of them don't exist, even though the authors are real sleep researchers.

A nursing student gets a reference with a DOI. When he opens doi.org, the DOI goes to an unrelated engineering paper, which tells him the citation was put together from plausible pieces.

A history student uses an AI search tool that links to a real journal article. When she reads it, the article discusses a different decade than the one the chatbot's summary claimed.

A graduate student asks the chatbot whether a citation it gave is real, and it confidently says yes. She checks the library catalog anyway and finds no such book.

风险与防护栏

  • 幻觉的事实可以悄悄地进入报告、支持流程或研究成果。

  • 及时的敏感性可能会在类似的请求中产生不一致的结果。

  • 如果访问控制薄弱,敏感文本数据可能会暴露。

实施路线图

  1. 在推出之前定义输出格式、语气和质量标准。

  2. 当准确性很重要时,请使用可信来源进行地面响应。

  3. 为高风险输出保留人工审查检查点。

  4. 跟踪故障模式并定期重新训练提示或工作流程。

不断探索

Free newsletter

Get the daily AI briefing

Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.

One email each weekday. Unsubscribe in one click. We never sell or share your address.

Test yourself

Take the Spotting Fake Citations from AI quiz

Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.

开始测验

Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation

常见问题

What is Spotting Fake Citations from AI?

Chatbots invent citations because they generate text that looks right, not text they have checked. A made-up reference can have real-sounding authors, a plausible journal and even a DOI. That matters because a fake or misattributed source in your paper can cost you the grade, damage your credibility, or in professional settings lead to formal sanctions.

Why can a chatbot produce a citation that doesn't exist?

Without a search step, the model produces text that fits the pattern of a citation, with no built-in check that the reference exists.

Which kind of bad citation does the guide call the hardest to spot?

When the paper exists and the details are right, only reading it shows that it doesn't say what the chatbot claimed.

What happened in Mata v. Avianca in 2023?

The lawyers filed a brief with court cases ChatGPT had made up, and the court sanctioned them.

When checking a DOI, what should you confirm after opening it through doi.org?

A made-up DOI may open a different, real paper, so you have to compare the title and authors, not just see that the link works.

What should you do if you can't verify a source?

If a source can't be verified, it doesn't belong in your work. The guide says to delete it.