DeepFake O'Meter
Siwei Lyu works with postdoctoral researcher Shan Jai in the UB Media Forensics Lab. Photo: Meredith Forrest Kulwicki
As it becomes easier and cheaper to use AI to create fake media, journalists and social media users are faced with the impossible task of trying to determine what to believe. Many turn to experts like CII Co-Director Siwei Lyu, who routinely obliges such requests. But the volume of requests can be overwhelming.
That inspired Lyu and his Media Forensics Lab to develop DeepFake-o-Meter, a free, open-source, web-based platform that combines several state-of-the-art, deepfake-detection algorithms.
“The goal is to bridge the gap between the public and the research community,” says Lyu, SUNY Empire Innovation Professor in the Department of Computer Science and Engineering, School of Engineering and Applied Sciences.
DeepFake-o-Meter (DFoM) will be updated as researchers develop new ways to detect inauthentic media, so it will remain cutting edge. Unlike commercially available products, DFoM allows users to run multiple detection algorithms, each providing a percentage likelihood that the content was AI generated.
The other feature that sets DFoM apart is transparency. “Other tools’ analysis may be accurate, but they do not disclose what algorithms they used to come to that conclusion and the user only sees one response, which could be biased,” Lyu says. “We’re trying to provide the maximum level of transparency and diversity with open-source codes from many different research groups.”
