Scientists from the University of Notre Dame, Indiana, have developed an AI-based early warning system to identify manipulated images, deepfake videos and disinformation online.
According to the scientists led by associate professor Tim Weninger, the scalable, automated system uses content-based image retrieval and applies computer vision-based techniques to root out political memes from multiple social networks.
“Memes are easy to create and even easier to share,” says Weninger.
“When it comes to political memes, these can be used to help get out the vote, but they can also be used to spread inaccurate information and cause harm.”
Weninger, along with assistant professor Walter Scheirer and their team, collected more than two million images and content from various sources on Twitter and Instagram related to the 2019 general election in Indonesia.
The results of that election, in which the left-leaning, centrist incumbent garnered a majority vote over the conservative, populist candidate, sparked a wave of violent protests that left eight people dead and hundreds injured.
Their study found both spontaneous and coordinated campaigns with the intent to influence the election and incite violence.
According to the scientists, those campaigns consisted of manipulated images exhibiting false claims and misrepresentation of incidents.
Moreover, logos belonging to legitimate news sources were also used on fabricated news stories with the intent to provoke citizens and supporters of both parties.
While the ramifications of such campaigns were evident in the case of the Indonesian general election, the threat to democratic elections in the West already exists.
Notre Dame is thus working on a new system that can flag manipulated content to prevent violence, and to warn journalists or election monitors of potential threats in real time.
The scientists contend that the system is scalable enough to provide users with tailored options for monitoring content.
They are also hoping to use their system to monitor the 2020 U.S. general election and have already begun collecting relevant data.
Image and content: Media Road/University of Notre Dame