Balancing Digital Safety between Risks, Rights, and Regulation
The Weizenbaum Institute (WI), the Bavarian Research Institute for Digital Transformation (bidt), the Center for Advanced Internet Studies (CAIS) and the Leibniz Institute for Media Research – Hans Bredow Institute (HBI) invite doctoral and postdoctoral researchers to present their work at the joint Digitalisation Research Seminar – DigiSem 2026 on: 21-22 September 2026 at WI in Berlin.
This year’s Digitalisation Research Seminar is dedicated to Digital Safety as a human-centered, interdisciplinary research field. We aim to bring together early career researchers who examine human and societal risks of digital technologies, including harm arising from bias, dis- and misinformation, loss of control, and institutional misalignment, as well as challenges in risk prevention for vulnerable groups, such as children.
Our goal is to promote interdisciplinary dialogue and networking among early career researchers, and to advance research that contributes to safer, more responsible, and socially sustainable digital systems. Constructive feedback and cross-cutting discussions are central to the seminar.
Keynotes
Prof. Dr. Lena Kästner (University Bayreuth): “AI and the Safety Paradox: Convenience, Control, or Compromise?”
Artificial intelligence is rapidly becoming embedded in everyday life, offering unprecedented convenience while also introducing new risks. One central concern in this context, especially pressing in the current political climate, is safety. On the one hand, AI may be deployed to enhance safety and protect our interests. On the other, it undoubtedly also creates significant safety hazards. The central question of Lena Kästner’s talk thus is how we shall navigate the ensuing safety paradox for AI, viz. that the very same technology simultaneously offers vast potential and vulnerabilities. Drawing on recent research from within the interdisciplinary Project “For the Greater Good? – Deepfakes in Criminal Investigations” (FoGG), Lena Kästner shall clarify how the paradox unfolds in practice. She explores how AI both enables novel forms of crime (e.g., AI-assisted fraud) and potentially enhances investigative capabilities of prosecutors. She shall examine under what conditions AI may and should legitimately be used and discuss the implications for modern society at large. Her central claim will be that effective AI governance must find a compromise between convenience and control to safeguard individuals and economic interests, protect democratic values, and remain robust in the face of evolving socio-technical risks.
Prof. Dr. Lena Kästner is professor for philosophy, computer science and artificial intelligence at University of Bayreuth and a PI on the project For the Greater Good? Deepfakes in Criminal Prosecution (FoGG) at bidt.
Dr. Jens Karschau (BSI): “Cyber-Resilience: Bridging Consumer Protection, Smart Hardware, and the AI Frontier”
In an increasingly interconnected digital landscape, information security has transitioned from a technical niche to a vital social challenge. We will explore the intersection of digital resilience and human behavior. We examine modern cyber threats, focusing on the psychological mechanics of phishing and social engineering, where communication itself is weaponized against the user. We further investigate the security of consumer ecosystems, specifically within the Smart Home as part of the BSI investigative series on consumer products. By analyzing hardware integrity and empirical product testing, we highlight systemic risks in everyday devices that often escape public scrutiny. Additionally, we briefly discuss the evolving role of Artificial Intelligence; shifting the focus from deepfakes toward AI’s capacity to automate and scale personalized threats. This talk argues that security is not merely a matter of robust code, but of understanding the socio-technical dynamics that shape our digital agency and security in an age of hyper-connectivity.
Dr. Jens Karschau is a civil servant at the department of digital consumer protection at the BSI (Federal Office for Information Security).
Programme
Monday, 21.09.2026
13:00 – 14:00: Arrival and Lunch
14:00 – 14:20: Welcome by the Organisers
14:20 – 15:20: Keynote talk: AI and the Safety Paradox: Convenience, Control, or Compromise? | Prof. Lena Kästner (University of Bayreuth/bidt-project PI)
15:20 – 15:45: Coffee Break
15:45 – 17:45: Session 1 – Risks of using AI on bias and decision making
- Cybersecurity as Individualized Responsibility: Organisational Security Communication and the Responsibilization of Digital Safety | Sonja Spitzer (University of Vienna, AUT)
- Governing Through Algorithms: The Normalisation of Algorithmic Surveillance in Democratic Systems | Suja Joseph (Universität Hildesheim)
- Balancing Utility and Fundamental Rights: A comparative analysis of Chinese and EU Law on Facial Recognition Technology | Yuying Wang (University of Münster)
- Out of scope, out of mind? The influence of unrelated AI advice on clinicians’ diagnostic reasoning during X-ray interpretation | Insa Schaffernak (TH Augsburg)
17:45 – 18:00: Refreshments
18:00 – 19:00: Poster Pitches & Poster Session
- Valerie Bohn (U of Goettingen): “The Illusion of Sentience – Ethical Implications and Implementation Guideline or Anthropomorphized Social AI and Robots“
- Dr. Mariia Pleskach (NTU of Ukraine):„Rethinking AI Fairness Beyond Explainability in EU AI Law as a Socio-Technical Accountability Approach“
- Yizhou Sun (FU Berlin): „The Neoliberal Code: Extractive Data and Extracted Consent“
- Eva Luise Knor (HBI / Leibniz-Institut für Medienforschung / U Hamburg): „Negotiating Risks: How Lay Users Respond to Generative AI Errors in Political Information Search“
- Dr. Jan Grenzebach (Federal Institute for Occupational Health and Safety, BAuA): „The Mechanical Turk 2.0: An Interface to Experiment with Human-AI Interaction Dynamics“
- Eva Ziegeltrum et al. (FAU Erlangen-Nürnberg, U Cambridge, U Zurich, U Tartu, U Vienna, U Navarra): „Mapping the Policy Landscape of Social Media Regulation to Improve Youth Mental Health: Towards a Rights-Informed Typology of Access, Usage, Design and Regulatory Environment “
- Corinna Canali (Weizenbaum Institute): „Systemic Obscenification as Governance “
- Leonhard Kestel & Christoph Kern (LMU München): „Counterfactual Latent Representations: A Neurosymbolic Approach“
19:00: Welcome Reception & Networking
Tuesday, 22.09.2026
9:00 – 10:30: Session 2 – Protection of vulnerable groups in the digital space
- tba
- How to protect children from AI systems risks? A critical approach | Dr. Silvia Corradi (University of Innsbruck)
- Participatory AI for Resource-Poor Languages: Correcting Moderation Failures through Survivor Knowledge | Pooja Lama, Himanshu Panday, Dridhata Silwal, Hameeda Syed & Saashi Windsor (Dignity in Difference)
10:30 – 11:00: Coffee Break
11:00 – 12:00: Keynote talk: Cyber-Resilience: Bridging Consumer Protection, Smart Hardware, and the AI Frontier | Dr. Jens Karschau (Bundesamt für Sicherheit in der Informationstechnik)
12:00 – 12:45: Lunch Break
12:45 – 14:15: Session 3 – Regulation of social media for human and societal safety
- Paid Attention: The Impact of Meta Ads in the 2025 German Federal Election Campaign | Jakob Berg, Prof. Johannes Breuer & Felix Schmidt (University of Regensburg, CAIS, gesis)
- Speaking the Law’s Language: Helping Digital Safety Researchers Navigate Regulatory Processes | Annabel Schultz & Dorian Tsolak (Media Authority of North Rhine-Westphalia)
- Surveillance Watermarking: Trading Privacy for AI Disclosure | Zachary Cooper (Weizenbaum Institute / Vrije Universiteit Amsterdam)
14:15 – 14.30: Closing Remarks
Programme of DigiSem 2026 as PDF
Registration
Register for DigiSem 2026 here: https://eveeno.com/digisem-2026
The deadline for the registration is 7 September, 2026.
Participation
We are happy to inform you that participation in the DigiSem is free of charge and that catering will be provided for participants throughout the event. Please note that travel expenses must be covered by participants themselves.
If you have any questions, please contact the responsible organisers as indicated below or have a look at the respective websites.
About the convening institutions:
bidt: The Bavarian Research Institute for Digital Transformation (bidt), an Institute within the Bavarian Academy of Sciences and Humanities, contributes to a better understanding of digital transformation’s developments and challenges. Thereby, we provide the foundations which will shape society’s digital future responsibly, for the common good.
CAIS: The Center for Advanced Internet Studies (CAIS) in North Rhine-Westphalia promotes the active shaping of the social, political, economic, and cultural changes that digitalization brings about. The Center sees itself as a place for innovative interdisciplinary research and as a source of inspiration for a critical public that wants to find agreement on models for a self-determined life in the digital society.
HBI: The research perspective of the Leibniz-Institute for Media Research | Hans-Bredow-Institut (HBI) focuses on media transformation and related structural changes of public communication. With its cross-medial, interdisciplinary and independent research, it combines basic research and transfer research, and thus, generates knowledge on issues relevant for politics, commerce and civil society.
WI: The Weizenbaum Institute for Networked Society (WI) is the German Internet Institute, a place of excellent research on the transformation and design processes of digital change. In the spirit of Joseph Weizenbaum, we research the necessary framework conditions, means and processes for individual and social self-determination in a networked society. We understand self-determination as a design principle that is central to the preservation of human dignity and democracy.