This text was published by IDMO, one of fifteen EDMO Hubs.
This content has been automatically translated to English using machine translation. As such, the translation may not be perfect and may contain errors or nuances that differ from the original text.
For the most accurate information, please refer to the original version.
For the most accurate information, please refer to the original version.
AI detection tools are known to sometimes misidentify AI-generated images as authentic, but a new NewsGuard audit of leading detection models has highlighted what may be an even more troubling weakness: three of the five tools tested are frequently fooled by real images.
