Ethics & SecurityNovember 16, 2026
0
ā„ļø

Fraunhofer's Deepfake Detector Flags Callers in One Second

A detector developed at Fraunhofer is reported to identify deepfake callers using face swap and voice cloning in under a second.

#Deepfakes#Fraunhofer#Voice Cloning#Security#Detection
Fraunhofer entlarvt Deepfake-Anrufer in Echtzeit
Share Article
šŸ”„ What happened Fraunhofer SIT demoed a deepfake detector at it-sa in Nuremberg in late October that flags face-swapped and voice-cloned callers in under a second. The proof-of-concept runs locally, shows a fraud probability score, and highlights suspicious image regions with a heatmap. šŸ’” Why it matters Scammers can now impersonate colleagues or bosses in real time, and a brief glitch in a video call rarely raises eyebrows. The detector processes everything on-device instead of routing calls to a server, but it demands so much compute that ordinary smartphones can't run it yet. ⚔ Our take A proof of concept that targets exactly the gap most companies ignore: video calls are an open door. Until this runs on every laptop, the best defense remains healthy suspicion of unexpected callers.
The title, summary and analysis of this item were produced automatically by an AI system and have not been editorially reviewed. They may contain errors, bias or omissions — when in doubt, read the linked original source.