AI Research Across 3 Countries Detects Fake News and Deepfakes

The Universitat Oberta de Catalunya (UOC) is collaborating with Polish and Japanese researchers from the Warsaw University of Technology and Okayama University to develop an innovative technology called DISSIMILAR that uses artificial intelligence and data concealment techniques to help users tell whether a video is real or a deepfake.

Social media nowadays is rife with the spread of fake news and misinformation, one of the most insidious forms of this being deepfakes. Deepfakes are videos which are edited in such a way, often with AI technology, to combine images, video footage, and audio to create a false or misleading videos.

The DISSIMILAR project works to combat this. Professor David Megías, KISON’s lead researcher and director of the IN3, said: “The project has two objectives: firstly, to provide content creators with tools to watermark their creations, thus making any modification easily detectable; and secondly, to offer social media users tools based on latest-generation signal processing and machine learning methods to detect fake digital content.”

Currently there are two types of tools used to detect fake news – automatic, based on machine learning, and much more commonly, platforms using human moderators. Social media sites such as Facebook and Twitter use human moderators and fact-checkers to tell whether content is real or fake, but that calls into question the danger of biases.

This human solution could be affected by “different biases” and encourage censorship, according to David Megías, who said: “We believe that an objective assessment based on technological tools might be a better option, provided that users have the last word on deciding, on the basis of a pre-evaluation, whether they can trust certain content or not”.

“That’s why we’ve opted to explore the concealment of information (watermarks), digital content forensics analysis techniques (to a great extent based on signal processing) and, it goes without saying, machine learning”, he continued.

Professor Megías explained that the fake news detection needed to be carried out with a combination of multiple different tools, both machine and human. The use of digital watermarking can be used to verify a media file “easily and automatically”. For example by confirming it had been distributed by a legitimate news industry or whether it had been posted from a brand new Twitter account. This too makes it so the user can easily tell if the content is fake.

DISSIMILAR will combine the development of digital watermarks with digital forensics to detect irregularities and distortions produced when modifying a file, using machine learning models to improve outcomes.

First-stage test studies are to take place in Catalonia, Poland, and Japan, with the researchers taking a holistic view on the perceptions and cultural components around spotting fake news. Professor Megías said: “Firstly, we want to find out how users interact with the news, what interests them, what media they consume, depending upon their interests, what they use as their basis to identify certain content as fake news and what they are prepared to do to check its truthfulness. If we can identify these things, it will make it easier for the technological tools we design to help prevent the propagation of fake news”.

The perceptions will be gauged in different cultural contexts, splitting the user groups between Catalonia, Poland, and Japan. Andrea Rosales, a CNSC researcher, said: “This is important because, for example, each country has governments and/or public authorities with greater or lesser degrees of credibility”.

“This has an impact on how news is followed and support for fake news: if I don’t believe in the word of the authorities, why should I pay any attention to the news coming from these sources? This could be seen during the COVID-19 crisis: in countries in which there was less trust in the public authorities, there was less respect for suggestions and rules on the handling of the pandemic and vaccination”, she added.

In the second stage, the users will participate in designing the tool, to “ensure the product will be well-received, easy to use and understandable”, said Andrea Rosales. The researchers envision social network platforms possibly integrating the software into their platforms to combat fake news. “If our experiments bear fruit, it would be great if they integrated these technologies”, concluded David Megías.