with ai
Researchers from Ghent University have developed a tool that undoes the watermark protection of Telenet’s .comdom app with the help of artificial intelligence. The researchers made the tool in a few days.

The researchers regret that Telenet did not pay sufficient attention to safety when developing the .comdom app. The provider released its .comdom app at the end of November to secure the mutual sending of spicy photos. The app places a watermark with the recipient’s details such as name and phone number on the image and makes the sender’s face unrecognizable. “This filter cannot even be removed by the experienced Photoshop user – or not without making the nude Photo unrecognizable,” Telenet wrote about his app . The app must prevent recipients from spreading the spicy photos, because they can be traced back to them.

Researchers Martijn Courteaux and Hannes Mareen of the IDLab-Media research team at UGent, however, managed to reverse the watermarks automatically in a short time. According to the makers, their software can ‘almost perfectly reconstruct’ the original photo, thus removing the data from the recipient. The researchers did this by finding out exactly how the watermark is built by the app, and by providing thousands of photos with such a watermark. They then trained a neural network on that collection of photos and the original variants. The trained algorithm was then able to recognize watermarks and build a photo as it would be.

According to the researchers, the characteristics of the watermarks were too simple and the power of ai algorithms had not been taken into account during development. They also note that the detection of the app to make faces unrecognizable does not always work well and does not work with multiple faces.

Telenet condom app watermark Ghent University