To counter the proliferation of deep fakes, Microsoft has released a new video authentication tool that can be used to analyze a still image or video to determine a percentage probability that a medium has been artificially manipulated.
In the case of a video, Microsoft stated that this percentage could be provided in real-time for each frame while the video is playing. Detects the blending line of deep, subtle grayscale or fading elements that may not be visible to the human eye.
Deepfakes or synthetic media can be photos, videos, or audio files that are manipulated by artificial intelligence (AI). Detecting Deep Forgeries Vital in Upcoming US Elections, According to Microsoft.
The technology was created using a public dataset from Face Forensic ++. Microsoft said it was tested with the DeepFake Detection Challenge dataset, which is believed to be the leading model. to train and test incoming counterfeit detection technologies.
“We anticipate that synthetic media generation methods will continue to grow in complexity. Since all AI detection methods have error rates, we must understand and be prepared to react to deep-seated fakes that pass through detection methods.” the company said in a blog post.
“In the long run, therefore, we must look for stronger ways to maintain and certify the authenticity of news articles and other media.”
With a few tools available, Microsoft has also introduced a new technology that can detect tampered content and ensure that the media you are watching is genuine.
The technology consists of two components. The first is a tool built into Microsoft Azure that enables a content producer to add digital hashes and certificates to content.
“Hashes and certificates live with content as metadata, wherever it is online,” Microsoft said.
The second is a reader that can be included in a browser extension that verifies certificates and compares hashes against matches to determine authenticity.
Microsoft has also partnered with the AI Foundation in its deep sham battle. As part of the partnership, both parties will provide video authenticators to organizations involved in the democratic process, including news organizations and political campaigns, through the foundation’s Reality Defender 2020 initiative.
The video authenticator will initially only be available through the initiative.
Another partnership with a consortium of media companies known as Project Origin is testing Microsoft’s authenticity technology. An initiative by several publishers and social media companies, the Trusted News Initiative, also agreed to work with Microsoft to test their technology.
The University of Washington, In-House Counterfeit Detection Sensitivity, and USA Today have also teamed up with Microsoft to improve media literacy.
“By improving media literacy, people can eliminate misinformation based on real events and manage the risks of cheap and deep fakes,” Microsoft said. “Knowing the media can allow us all to think critically about the media context and become more engaged citizens while enjoying satire and parody.”
As part of this partnership, there will be a public service announcement campaign that will invite people to “pause for reflection” and verify that the information comes from a reputable news organization before sharing or promoting it. on social media before the elections.
The parties also launched a questionnaire for American voters to learn more about synthetic media.