Public Interest AI
The use of artificial intelligence for the public interest is repeatedly formulated as a central goal in political AI strategies, guidelines and government funding measures, as well as in projects from commercial enterprises. However, it is still unclear whether and, above all, how the public interest orientation of AI applications can be realised and how it can be checked. The “Public Interest AI” research group is therefore developing an operationalisable analysis procedure that will allow Public Interest to be defined in an AI-specific way and tested in technical prototypes. With this practice-oriented research approach, the research group is developing an exemplary catalog of requirements for assessing to what extent socially oriented AI projects serve the public interest. As a result, this can function both as an important contribution to the discourse on AI development in Europe and as practical implementation guidelines for AI developers.
PhD projects
In three PhD projects, different aspects of the concept of AI for social good are being studied.
The first PhD project focuses on processes for the validation of public AI systems by public actors and related deliberative design processes. The format of design patterns is central to the project, which makes it possible to capture and make available transferable patterns for public good-oriented AI applications. Qualitative case studies will be conducted to shape the content of the design patterns.
The second project investigates the role of fact-checks in public discourse and how they can be supported by computer-based methods. To this end, a theoretical framework on claims that potentially contain disinformation and a machine learning model that automatically classifies these types of claims will be developed. In addition, it is planned to empirically investigate the reception and impact of fact checks.
The third PhD project deals with the automated simplification of German-language texts. Besides the purely scientific impact, the aim of the project is to minimise the exclusion factor for less literate citizens. The focus of the technical work is on individual sub-components that should contribute to the goal of full automation. The first sub-component in the project relates to the relationship between summaries and simplifications.
| Duration | October 2020 – September 2024 |
| Sponsor | Federal Ministry of Education and Research | FKZ 01IS20058 |
Theresa Züger, Dr.
New Open Talk Format
In our "Open Talk" series, we invite experts from the field of public good-oriented AI. With their different perspectives from research and practice, they help us to develop holistic answers on the use of artificial intelligence for the public interest.
Christopher RichterStudent Assistant: Public Interest AI
Daniel Cameron PothmannProject Assistant: Knowledge Transfer | Public Interest AI
Ferdinand Bálint DorendorfStudent Assistant: AI & Society Lab
Freya HewettResearcher: AI & Society Lab
Hadi Asghari, Dr.Researcher: AI & Society Lab
Irina KühnleinStudent Assistant: AI & Society Lab
Johanna Teresa WallenbornAssociate Researcher: Knowledge & Society and SET
Judith FaßbenderResearcher: AI & Society Lab
Sami NennoResearcher: AI & Society Lab
Theresa Züger, Dr.Research Group Lead: Public Interest AI | AI & Society Lab
-
Jakob Stolberg-LarsenFormer Researcher: AI & Society Lab
Journal articles and conference proceedings
Asghari, H., Stolberg-Larsen, J., & Züger, T. (2022). Approximating Accessibility of Regions from Incomplete Volunteered Data. CHI EA '22: Extended Abstracts of the 2022 CHI Conference on Human Factors in Computing Systems, 1-6. DOI: 10.1145/3491101.3519706 Publication details
Züger, T., & Asghari, H. (2022). AI for the public. How public interest theory shifts the discourse on AI. AI & Society. DOI: 10.1007/s00146-022-01480-5 Publication details
Hewett, F. & Stede, M. (2022). Extractive Summarisation for German-language Data: A Text-level Approach with Discourse Features. Proceedings of the 29th International Conference on Computational Linguistics (COLING), 756-765. Publication details
Asghari, H. & Hewett, F. (2022). HIIG at GermEval 2022: Best of Both Worlds Ensemble for Automatic Text Complexity Assessment. Proceedings of the GermEval 2022 Workshop on Text Complexity Assessment of German Text, 15–20. Publication details
Asghari, H., van Biemen, Thomas, & Warnier, Martijn (2021). Amplifying Privacy: Scaling Up Transparency Research Through Delegated Access Requests. 5th Workshop on Technology and Consumer Protection (ConPro ’21). Publication details
Mahieu, R., Asghari, H., Parsons, C., van Hoboken, J., Crete-Nishihata, M., Hilts, A., & Anstis, S. (2021). Measuring the Brussels Effect through Access Requests: Has the European General Data Protection Regulation Influenced the Data Protection Rights of Canadian Citizens? Journal of Information Policy, 11, 301-349. DOI: 10.5325/jinfopoli.11.2021.0301 Publication details
Hewett, F., & Stede, M. (2021). Automatically evaluating the conceptual complexity of German texts. Proceedings of the 17th Conference on Natural Language Processing (KONVENS 2021), 228–234. Publication details
Book contributions and chapters
Züger, T. (2021). The world we think is the world we get. In T. Bazzichelli, Whistleblowing for Change. Exposing Systems of Power & Injustice. Berlin, Germany: transcript. Publication details
Working paper
Asghari, H., Birner, N., Burchardt, A., Dicks, D., Fassbender, J., Feldhus, N., Hewett, F., Hofmann, V., Kettemann, M. C., Schulz, W., Simon, J., Stolberg-Larsen, J., & Züger, T. (2022). What to explain when explaining is difficult. An interdisciplinary primer on XAI and meaningful information in automated decision-making. Humboldt Institute for Internet and Society. DOI: 10.5281/zenodo.6375784 Publication details
Züger, T., Faßbender, J., Kuper, F., Nenno, S., Katzy-Reinshagen, A., & Kühnlein, I. (2022). Civic Coding: Grundlagen und empirische Einblicke zur Unterstützung gemeinwohlorientierter KI. Civic Coding Initiative. Publication details
Other publications
Fassbender, J. (2022). Why explainable AI needs such a thing as Society. Digital society blog. Publication details
Nenno, S., & Richter, C. (2022). Sustainable AI – How environmentally friendly is AI really? Digital society blog. Publication details
Züger, T. (2022). A lion for sustainable AI: How to support a new standard for sustainability reporting? Digital society blog. Publication details
Nenno, S. (2021). Siri’s evil sister. When the Dutch public service steals your data. Digital society blog. Publication details
Hewett, F., & Stede, M. (2021). Lexica corpus, . Publication details
Hewett, F., & Nenno, S. (2021). How to identify bias in Natural Language Processing. Digital society blog. Publication details
Nenno, S. (2021). How to find the right survey tool:problems and proposals. Elephant in the Lab. Publication details
Lectures and presentations
Seminar: Demystifying AI: Typologies to map AI-systems and datasets.. KISD - Köln International School of Design. KISD - Köln International School of Design, Cologne, Germany: 18.11.2022Judith Fassbender
Public Interest AIUNESCO Fachausschuss Kommunikation und Information. German Comission for UNESCO. UNESCO Fachausschuss Kommunikation und Information, Brussels, Belgium: 15.11.2022 Further information
Theresa Züger
Was bedeutet Transparenz für KI-Systeme?ZVKI Fachforum. iRights.Lab. Online, Berlin, Germany: 10.11.2022 Further information
Theresa Züger
Digital tools for text simplificationSeminar “Sprache als Schwelle zur Kunst? Kunstvermittlung und Konzepte einfacher und leichter Sprache”. Kunsthalle Osnabrück. Online, Osnabrück, Germany: 09.11.2022
Freya Hewett
Civic Coding – KI für das Gemeinwohl nutzenCivic Coding – KI für das Gemeinwohl nutzen. Bundesministerium für Arbeit und Soziales. Impact HUB Berlin, Berlin, Germany: 18.10.2022 Further information
Theresa Züger, Judith Fassbender, Freia Kuper
Design Patterns – Prozessformate für eine verantwortungsvolle Umsetzung von KI-SystemenKI-Lab. The Policy Lab Digital, Work & Society of the Federal Ministry of Labour and Social Affairs. KI-Lab of the Policy Lab Digital, Work & Society of the Federal Ministry of Labour and Social Affairs, Berlin, Germany: 30.09.2022
Judith Fassbender
Poster Pitch: Are AI standardisation processes inclusive enough to deliver tech that works for everyone?European Think Tank Conference 2022. German Council on Foreign Relations (DGAP). European Think Tank Conference 2022, Berlin, Germany: 20.09.2022 Further information
Hadi Asghari
Approximating Accessibility of Regions from Incomplete Volunteered DataCHI EA '22. Association of Computing Machinery. Online, New Orleans, USA: 03.05.2022 Further information
Hadi Asghari
Seminar: Programmatische Autonomie: Künstliche Intelligenz und selbstfahrende Autos (Session: Lecture: Künstliche Intelligenz?). University of Potsdam: Prof. Dr. Jan Distelmeyer. Potsdam University, Potsdam, Germany: 03.05.2022Theresa Züger
PhD Workshop on Artificial Intelligence and Machine Learning research and DemocracyPhD Workshop on Artificial Intelligence and Machine Learning research and Democracy. University of Tübingen. University of Tübingen, Tübingen, Germany: 02.04.2022 Further information
Judith Fassbender
Particles of a whole: design patterns for transparent and auditable AI-systemsUbiComp '21: The 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing (Session: Particles of a whole: design patterns for transparent and auditable AI-systems). UbiComp. Online, New York, USA: 25.09.2021 Further information
Judith Faßbender
Automatically evaluating the conceptual complexity of German textsKONVENS 2021 (Session: Automatically evaluating the conceptual complexity of German texts). German Society for Computational Linguistics and Language Technology, the Special Interest Group on Computational Linguistics of the German Linguistic Society and the Austrian Society for Artificial Intelligence. Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany: 09.09.2021 Further information
Freya Hewett
AI with Volunteering CommunitiesThe 3rd Workshop on Artificial Intelligence for Social Good (Session: AI with Volunteering Communities). International Joint Conferences on Artificial Intelligence, Montreal, Canada: 21.08.2021 Further information
Stolberg-Larsen, J.
Amplifying Privacy: Scaling Up Transparency Research Through Delegated Access Requests5th Workshop on Technology and Consumer Protection (ConPro ’21) (Session: Amplifying Privacy: Scaling Up Transparency Research Through Delegated Access Requests). ConPro ’21. Online, San Francisco, USA: 27.05.2021 Further information
Hadi Asghari
AI in the public interest: What does this mean and how can be build it?Gesellschaft für Informatik: #KICamp21. Gesellschaft für Informatik e.V.. Online, Berlin, Germany: 24.02.2021 Further information
Theresa Züger, Hadi Asghari, Freya Hewett, Judith Fassbender, Jakob Stolberg-Larsen
Panels
Künstliche Intelligenz + Gemeinwohl = ?Die dima und du! Dein Zukunftsassistent lädt ... ein!. Projektmitarbeiterin der Digitalen Manufaktur (dima) Stadt Aschaffenburg. JUKUZ - Jugendkulturzentrum Aschaffenburg, Aschaffenburg, Germany: 28.10.2022 Further information
Theresa Züger
The Future of War: Drone Surveillance, Killer Robots & Lethal AI WeaponsDNL26: The Kill Cloud: Networked Warfare, Drone & AI. Disruption Network Lab. Kunstquartier Bethanien, Berlin, Germany: 25.03.2022 Further information
Theresa Züger
Wie vermitteln wir Wissen über KI an die Bevölkerung?Auftaktveranstaltung: Zentrum für vertrauenswürdige Künstliche Intelligenz. iRights.Lab, Berlin, Germany: 07.12.2021 Further information
Theresa Züger
Young activists & human rights in a digital ageFundamental Rights Forum. Online, Fundamental Rights Forum, Warsaw, Poland: 11.10.2021 Further information
Theresa Züger
Gemeinwohlorientiert = Gemeinnützig? Unternehmen und gemeinwohlorientierte KIre:publica 21. Im Internet, Berlin, Germany: 21.05.2021 Further information
Theresa Züger
Moderation of workshops and panels
Automatisation and autonomyInfrastructures of Autonomy. Humboldt Institute for Internet and Society, Brandenburg Center for Media Studies. Humboldt Institute for Internet and Society, Berlin, Germany: 24.11.2022 Further information
Hadi Asghari
Organisation of events
Sustainable AI IIwith attending Vip: Lynn Kaack. 14.06.2022. Humboldt Institute for Internet and Society, Berlin, Germany (National)
Sami Nenno, Theresa Züger
AI and the Public Interest13.06.2022. Humboldt Institute for Internet and Society, Berlin, Germany (International) Further information
Hadi Asghari, Sami Nenno, Freya Hewett, Judith Faßbender, Theresa Züger
AI & Society Lab (AI & Society Lab)Study week of the ZHAW: "Der Mensch im digitalen Zeitalter". 05.05.2022. Humboldt Institute for Internet and Society, Berlin, Germany. Co-Organised by: ZHAW (International)
Sami Nenno, Judith Faßbender, Theresa Züger
Sustainable AIwith attending Vip: Friederike Rohde. 29.04.2022. Humboldt Institute for Internet and Society, Berlin, Germany (National)
Sami Nenno, Theresa Züger
Open Talk with Alex C. Engler: Algorithmic Services in the German GovernmentOpen Talk. with attending Vip: Alex C. Engler. 01.04.2022. online, Berlin, Germany (International) Further information
Open Talk with Anna Mätzener: Automatic Decision Making Systems
Open Talk Series. with attending Vip: Anna Mätzener. 20.03.2022. online, Berlin, Germany (International) Further information
Open Talk with Dr. Stephan Dreyer: Public Service media & Public Value
Open Talk Series. with attending Vip: Dr. Stephan Dreyer. 11.03.2022. online, Berlin, Germany (International) Further information
Open Talk with Julia Gundlach: Algorithms for the common good
Open Talk Series. with attending Vip: Julia Gundlach. 10.02.2022. online, Berlin, Germany (International) Further information
Research Clinic: Explainable AI
From 08.09.2021 to 12.09.2021. Humboldt Institute for Internet and Society, Berlin, Germany (National) Further information
Digitaler Salon: KI – Die Letzte räumt das Internet auf
31.03.2021. Livestream, Berlin, Deutschland (National) Further information
Natasha Vukajlovic, Philip Meier, Christian Grauvogel
Nominations
Research Clinic Europe – Platform://Democracy.Theresa Züger
Think Tank School of the German Council on Foreign Relations.Hadi Asghari, Theresa Züger