Research on social impacts of artificial intelligence funded with 850,000 euros

Press release from June 17, 2020


AI plays an increasingly important role in many areas of life – and raises fundamental legal and ethical questions. The Baden-Württemberg Foundation has now launched a new funding program on the topic of “Responsible Artificial Intelligence” and has awarded ten projects – three of them from scientists at the University of Mannheim. Research projects that examine the interaction of AI technologies and society are funded.

“The fact that three out of ten funded projects are located at the University of Mannheim is a great success for us,” said Prof. Dr. Heiko Paulheim, who heads two of the successful research projects. “This confirms Mannheim's central role in the areas of data science and artificial intelligence.” The interdisciplinary projects form a joint cluster in the Mannheim Center for Data Science, four faculties of the university are involved.

Illegal online price fixing
As part of the project “Antitrust-Compliant AI (KarekoKI)”, Prof. Dr. Thomas Fetzer and Prof. Dr. Heiko Paulheim, holder of the chair for data science, develop legal framework and strategies for the technical prevention of AI-based price fixing. The background: AI systems are increasingly being used on online platforms such as Amazon, which monitor the prices of competitors and set their own prices on this basis. “This raises legal and technical questions about how AI should deal with the potential of illegal price fixing,” says Prof. Dr. Thomas Fetzer, Chair of Public Law, Regulatory Law and Tax Law. The team receives funding of 383,000 euros for its research.

Social polarization through online news
The project “Responsible News Recommender Systems” (ReNewRS) by Prof. Dr. Heiko Paulheim and Dr. Philipp Müller from the Institute for Media and Communication Science in collaboration with Prof. Dr. Harald Sack (FIZ Karlsruhe) and Prof. Dr. Christof Weinhardt (KIT Karlsruhe) receives almost 700,000 euros, around half of which goes to the University of Mannheim. Message recommendation systems are used in social media platforms and on online news portals and recommend messages to users based on their previous use of the platform. The researchers want to investigate the effects of such systems on social polarization processes, political participation and prosocial behavior. For this purpose, different versions of such recommendation systems are to be developed in the project and their effects will be tested in online and laboratory studies.

Fairness and discrimination in automated decision-making processes
The project “Fairness in Automated Decision-Making (Fair ADM)” by Prof. Dr. Frauke Kreuter, Dr. Ruben Bach and Dr. Christoph Kern from the Chair of Statistics and Social Science Methodology, deals with discrimination and fairness of algorithm-based decision-making processes (automated decision-making, ADM) in the German public sector. “While ADM systems optimize bureaucratic processes through automation, their use also raises new social and ethical questions,” said Prof. Dr. Frauke Kreuter. It is feared that ADM could increase existing social discrimination. ADM systems are already used in the US context, for example, to assess the risk of relapse of defendants in the context of court proceedings. A particularly sensitive field of application of ADM in the European context is the assessment of the job market opportunities of job seekers, e.g. for the allocation of further training resources, which has currently been proposed by the Austrian Labor Market Service (AMS). There is a risk that sensitive characteristics such as gender, age or marital status will be used in algorithmic decision-making and thus influence the distribution of resources. In order to shed more light on this and to empirically investigate methods for correcting unfair algorithms, the project is developing and evaluating an ADM based on administrative labor market data. This research is supported with 171,000 euros.

Link to the press release of the Baden-Württemberg Foundation: https://www.bwstiftung.de/news-presse/pressemitteilungen/pressemitteilungen/programm-fuer-verantwortliche-kuenstliche-intelligenz-gestartet/?no_cache=1

Contact:
Prof. Dr. Thomas Fetzer
University of Mannheim
Department of Law
Tel .: +49 621 181-1438
Email: lsfetzer (at) mail.uni-mannheim.de

Prof. Dr. Frauke Kreuter
Chair of Statistics and Methodology
Co-director of the Mannheim Center for Data Science
University of Mannheim
Tel. +49 621 181-2003
Email: frauke.kreuter (at) uni-mannheim.de

Prof. Dr. Heiko Paulheim
Chair of Data Science
University of Mannheim
Tel. +49 621 181-2652
Email: heiko (at) informatik.uni-mannheim.de

Yvonne Kaul
Research communication
University of Mannheim
Tel. +49 174 3146512
Email: kaul (at) uni-mannheim.de