New Tool Can Tell If Something Is AI-Written With 99% Accuracy

Summary

A new study has found that an AI detector developed by the University of Kansas can detect AI-generated content in academic papers with a 99% accuracy rate. The model had a 100% accuracy rate of identifying human-created articles from AI-generated ones, and a 92% accuracy rate for identifying specific paragraphs within the text. AI detectors have not proven to be 100% accurate, and there are already many existing programs and services that promise to identify AI-written content. OpenAI's Classifier has a success rate of around 26%, and other programs like Winston AI and Quill and CommonLit's AI Writing Check have accuracy rates between 80% and 99%.

Q&As

AI Comments

👍 This new study is incredibly impressive in its ability to detect AI-generated content with a 99% accuracy rate. It's great to see researchers using AI for the betterment of society.

👎 It's concerning to hear that some universities are falsely accusing students of cheating by using AI. This could be incredibly damaging for students who are innocent, and is a reminder of the need to be cautious when using AI detection tools.

AI Discussion

Me: It's about a new AI detector that can detect AI-generated content in academic papers with a 99% accuracy. It's one of the few detectors on the market that can do this specifically for academic writing.

Friend: Wow, that's really impressive! What are the implications of this?

Me: Well, it has the potential to make it easier to differentiate between human-written and AI-generated content, which has already become a problem in some schools. The tool could also help ensure that no one is unfairly accused of using AI for their assignments when they haven't. On the other hand, there are some concerns that this could lead to people relying too heavily on AI-generated content in their writing, which would make it harder to recognize originality and creativity.

Action items

• Research existing programs and services that promise to identify AI-written content.
• Learn more about OpenAI’s Classifier and its accuracy rate for detecting AI-generated text.
• Explore the potential of using AI detectors to help detect plagiarism in academic papers.

Technical terms

AI-Written

Content that has been written by artificial intelligence (AI) software.

ChatGPT

An AI-powered chatbot created by OpenAI that can generate text from a given prompt.

AI Detector

A tool that can detect AI-generated content in academic papers.

Cell Reports Physical Science

A scientific journal that published a report on the AI detector.

Turnitin

A plagiarism detection software that can also detect AI-generated content.

GPT-3

A language model created by OpenAI that can generate text from a given prompt.

GPT-4

A language model created by OpenAI that is similar to GPT-3.

Jasper AI

An AI-powered text generation tool.

Winston AI

An AI-powered text detection tool.

OpenAI Classifier

An AI-powered tool created by OpenAI that can distinguish between AI-written and human-written text.

Quill and CommonLit

Companies that developed AI Writing Check, a tool to help teachers check for AI-created work in assignments.

Similar articles

0.95193183 New Tool Can Tell If Something Is AI-Written With 99% Accuracy

0.90888995 Teachers Are Absolutely Loving The Student Who Made A Tool That Shows If Your Essay Was Written By AI

0.8956027 A fake news frenzy: why ChatGPT could be disastrous for truth in journalism

0.892641 Researchers Poke Holes in Safety Controls of ChatGPT and Other Chatbots

0.88939404 ChatGPT is making up fake Guardian articles. Here’s how we’re responding