GLTR is a cutting-edge tool created by the MIT-IBM Watson AI lab and HarvardNLP to identify text that was generated automatically using forensic analysis. It uses a variety of statistical techniques to spot text creation artefacts like repeating patterns and strange word selections. Anyone can use GLTR to determine whether or not a textual passage was produced by an AI model, making it a useful tool for identifying false information and other types of misinformation.
