This year, an extensive study will be launched at the University of Helsinki focused on automated trolling and dissemination of disinformation.
The name of the project to be carried out by a research group headed by Antti Ukkonen, an Academy of Finland research fellow, is “Automated trolling and fake news generation in future social media: computational and empirical investigations of the threat and its implications”. The project will examine mechanisms through which online discussions and news can be interfered with utilising artificial intelligence.
The study received €700,000 in funding from the Media and Society programme of the Academy of Finland, from the beginning of 2019 to the end of 2022.
Fake websites through automation
The study combines the qualitative analysis of trolling and disinformation with modern machine learning techniques.
According to Ukkonen, who is working at the Department of Computer Science, University of Helsinki, the automated production of text and images through artificial intelligence technologies has taken vast strides in the last decade.
“In the future, it may be possible that even more advanced versions of these technologies could be used, among other things, for the large-scale automation of trolling or to generate entire websites comprised of nothing but more or less artificial news articles, illustrations and reader discussions,” Ukkonen explains.
The group aims to study computational methods that can help individuals and communities better prepare for situations where deciphering the authenticity of websites and the motivations of discussion participants becomes increasingly difficult.
“In addition to specialists in machine learning, our research group comprises Antti Salovaara, an expert in human-computer interaction, as well as Matti Nelimarkka and Pihla Toivanen, specialists in computational social science, while we also aim to recruit a postdoctoral researcher of qualitative research methods focused on interaction,” says Ukkonen.