Under Alphabet, Jigsaw is tasked with using technology to tackle global security challenges. Its latest project focuses on combating disinformation by detecting manipulated images with a new Assembler tool.
In the works since 2016, Assembler is an “early stage” experimental platform by Jigsaw to identify manipulated images. It hopes to reduce the various tools that journalists and fact-checkers use today by bringing “together multiple image manipulation detectors from various academics.”
Individually, these detectors can identify very specific types of manipulation — such as copy-paste or manipulations to image brightness. Assembled together, they begin to create a comprehensive assessment of whether an image has been manipulated in any way.
These models combine to determine the probability of manipulation in an image. Jigsaw, along with Google Research, created two new detectors for Assembler focused on deepfakes and to compile all the manipulated image results:
- The first is the StyleGAN detector to specifically address deepfakes. This detector uses machine learning to differentiate between images of real people from deepfake images produced by the StyleGAN deepfake architecture.
- Our second model, the ensemble model, is trained using combined signals from each of the individual detectors, allowing it to analyze an image for multiple types of manipulation simultaneously. Because the ensemble model can identify multiple image manipulation types, the results are, on average, more accurate than any individual detector.
Meanwhile, Jigsaw also announced The Current — a research publication that “illuminates complex problems through an interdisciplinary approach.” The first issue is on “exposing the architecture of disinformation campaigns.”
Our first issue is, as you might have guessed, all about disinformation — exploring the architecture of disinformation campaigns, the tactics and technology used, and how new technology is being used to detect and stop disinformation campaigns.
More about Jigsaw:
FTC: We use income earning auto affiliate links. More.
Check out 9to5Google on YouTube for more news: