Content generated by AI such as ChatGPT contains errors such as common sense and reasoning, but they can still be identified.

News on March 18, at the Association for the Advancement of Artificial Intelligence conference held in February this year, The results of a new study showed that it is possible to determine whether content is generated by ChatGPT or not. Written by humans.

A research team from the School of Engineering and Applied Sciences at the University of Pennsylvania recently launched the largest artificial intelligence detection campaign in history, based on web page training created by the university. Game​​Real or Fake Text? ​​, collecting relevant data for training so that the AI ​​can determine whether the content was generated by ChatGPT or written by humans.

The co-author of the study, Dr. Liam Dugan, explained this. The translation of IT House is as follows:

Today’s artificial intelligence can already generate very smooth and very Grammatical text. But artificial intelligence makes mistakes.

We have shown that machines make errors such as common sense errors, correlation errors, reasoning errors, and logic errors, and we have figured out how to detect these errors.

People are anxious about artificial intelligence. Our research may alleviate some of that anxiety. I think currently AI can help us write more imaginative and interesting texts, best suited for creative collaboration. But the use of AI in news reporting, academic papers, or legal advice is still so bad that we can’t be sure of its veracity.

The report content is as follows:​​https://www.php.cn/link/fc5b3186f1cf0daece964f78259b7ba0​​

The above is the detailed content of Content generated by AI such as ChatGPT contains errors such as common sense and reasoning, but they can still be identified.. For more information, please follow other related articles on the PHP Chinese website!