Journalists, screenwriters and college professors are among widening groups of people who are concerned about eventually\u00a0losing their livelihoods to artificial intelligence programs like ChatGPT, which can produce copy faster and possibly better than humans.<\/p>\n
But one entrepreneur is pursuing technology to make it easier to distinguish between text written by people and that composed by a machine.<\/p>\n
Edward Tian<\/a><\/strong>, a 22-year-old Princeton University student studying computer science and journalism, developed an app called\u00a0GPTZero<\/a><\/strong>\u00a0to deter the misuse of the viral chatbot ChatGPT in classrooms. The app has racked up 1.2 million registered users since January.<\/p>\n He\u2019s now launching a new program called Origin aimed at \u201csaving journalism,\u201d by distinguishing AI-generated disinformation from fact in online media.<\/p>\n Tian has secured $3.5 million in funding co-led by Uncork Capital and Neo Capital<\/a><\/strong>, with tech investors including Emad Mostaque, chief executive officer of Stability AI Ltd, and Jack Altman.<\/p>\n GPTZero analyzes the randomness of text, known as perplexity, and the uniformity of this randomness within the text \u2014 called burstiness \u2014 to identify when AI is being used. The tool has an accuracy rate of 99% for human text and 85% on AI text, according to the company.<\/p>\n The 10-person team now wants to empower journalism and is talking with large media organizations like the BBC and industry executives including New York Times former Chief Executive Officer Mark Thompson, to discuss partnerships for AI detection and analysis. The company also sees its technology for use in fields of trust-and-safety, government, copyright, finance, law and more.<\/p>\n \u201cWe believe we can get the smartest people working on AI detection in a room together,\u201d said Tian. \u201cThe field of detection is so new and we believe it deserves more attention and support.\u201d<\/p>\n Open AI, the company behind ChatGPT, has launched an AI\u00a0text classifier<\/a><\/strong>\u00a0to detect machine-generated content, but it\u2019s far from foolproof. The tool correctly identifies only 26% of AI-written text as \u201clikely AI-written,\u201d while incorrectly labeling human-written text as AI-written 9% of the time. The classifier also works \u201csignificantly worse\u201d in languages other than English and is \u201cunreliable\u201d on code and shorter texts. For inputs that are very different from text in the tool\u2019s training set, the classifier could also be wrong, according to OpenAI.<\/p>\n \u201cOur classifier has a number of important limitations,\u201d the company acknowledges on the website. \u201cIt should not be used as a primary decision-making tool, but instead as a complement to other methods of determining the source of a piece of text.\u201d<\/p>\n The lack of reliability of the detection tool poses a\u00a0dilemma for educators. Even if a teacher finds a suspicious article from a student that\u2019s flagged with a 70% likelihood of being AI-generated, as long as the accuracy of those detection tools isn\u2019t 100%, it\u2019s very hard for teachers to take decisive action.<\/p>\n \u201cI don\u2019t think we know what to do with a flag that says there might be an issue,\u201d said\u00a0Jack Cushman<\/a><\/strong>, director of the Harvard Library Innovative Lab, which explores topics such as the impact of the internet. \u201cAll you can do at that point is talk with a student and say you might have committed academic dishonesty according to this tool.\u201d<\/p>\n Meanwhile, the definition of plagiarism is also evolving with the emergence of AI. \u201cIt is going to challenge the whole notion of academic honesty because sometimes having a tool that recommends a sentence or two or help with citations is going to be legitimate in the same way as using calculator to do math work,\u201d he said. \u201cThe best answer is you shouldn\u2019t let it write the whole thing.\u201d<\/p>\nLack of Tools<\/h3>\n
Rise of Deepfakes<\/h3>\n