The UI is bare-bones but you’ll get the idea. Drag or upload an image and our tool will display the probabilities with which it thinks that the image might be AI-generated or not. If you want, you can click “No, it’s AI” to confirm that the image was AI-generated, or “No, it’s real” to confirm that the image was not AI-generated.
Why we’re working on this: as AI-generated images continue to blur the line between real and artificial and their adoption and quality rises, so too does the risk for fraud and misinformation. Not being able to trust what you see online threatens whatever level of "realness" or authenticity online material has. Companies like dating apps, news sites, and trust and safety teams have a growing need to distinguish AI-generated images from authentic ones.
The models we built are trained on various architectures, such as Dalle-3, Midjourney, and SDXL, with continuous integration of data from the latest AI image generators. Our technology can detect deepfakes and verify user profile images, documents, IDs, or media images. Additionally, it can detect fake or counterfeit products, services, or experiences being marketed on e-commerce platforms.
We hope it’s fun and would be very interested in any cases it gets wrong, as well as whatever else you’d like to ask or say!
Show HN: Fake or real? Try our AI image detector
trial.nuanced.dev
