New tools help identify AI-generated text and media
-
GPTZero helps teachers and professors identify AI-generated text by analyzing documents and highlighting text written by AI models vs humans. It's meant to help, not punish students.
-
Copyleaks AI Content Detector can identify AI-generated text with 99.1% accuracy across multiple languages. Users can paste text to have it analyzed.
-
Google's SynthID embeds imperceptible digital watermarks to identify AI-generated images. Tips watch for extra fingers, blurry details, uncanny/overly contrasted images.
-
Deepware Scanner scans videos to detect deep fakes and other AI manipulations. It accepts requests to analyze suspicious videos.
-
Other tools mentioned Sentinel, Reality Defender, WeVerify, Microsoft’s Video Authenticator.