Theses

This page contains automatically translated content.

Assignments of topics and supervision

...

 

Procedure

As part of the accompanying seminar, students will give three presentations, each lasting 10 to 20 minutes. These include a presentation on topic selection, the presentation of the proposal, and an interim or final presentation.

Depending on the degree program and examination regulations, a final colloquium (a 30- to 60-minute presentation) may be required.

Overview of the Topic Areas

Contact: Eva Späthe, eva.spaethe[at]uni-kassel[dot]de

Motivation and Relevance

  • Data protection is regulated by law, organizational policies, and technical standards; the requirements are constantly evolving alongside new technologies and areas of application.
  • In addition to legal requirements (e.g., the GDPR), governance structures, standards, and certifications shape practical implementation in organizations and institutions.
  • Relevant issues include the effectiveness, enforcement, and interoperability of regulatory frameworks, as well as their impact on society, businesses, government, and individuals.

Potential Research Areas

  • Implementation and effectiveness of regulatory requirements in specific industries, organizational types, or societal contexts
  • The role of standards, certifications, and codes of conduct
  • Governance structures and responsibilities in corporate data protection
  • The tension between data protection and new technologies (e.g., AI, cloud computing, tracking)

Possible research methods

  • Conceptual studies (systematic literature review)
  • Empirical studies (interviews, surveys, experiments)

Introductory Reading

  • Hinterleitner, M., Knill, C., & Steinebach, Y. (2024). The growth of policies, rules, and regulations: A review of the literature and research agenda. Regulation & Governance, 18(4), 1330–1348. https://doi.org/10.1111/rego.12511
  • Smith, H. J., Dinev, T., & Xu, H. (2011). Information privacy research: An interdisciplinary review. MIS Quarterly, 35(4), 989–1015. https://doi.org/10.2307/41409970
  • Späthe, E., Danylak, P., Lins, S., & Sunyaev, A. (2025). Demonstrating data protection efforts in companies. Data Protection and Data Security – DuD, 49(10), 649–656.
  • Taeihagh, A., Ramesh, M., & Howlett, M. (2021). Assessing the regulatory challenges of emerging disruptive technologies. Regulation & Governance, 15(4), 1009–1019. https://doi.org/10.1111/rego.12392
  • Zaguir, N. A., de Magalhães, G. H., & de Mesquita Spinola, M. (2024). Challenges and enablers for GDPR compliance: Systematic literature review and future research directions. IEEE Access, 12, 81608–81630. https://doi.org/10.1109/ACCESS.2024.3406724

Contact: Eva Späthe, eva.spaethe[at]uni-kassel[dot]de

Background on the topic:

  • Increasing digitalization, particularly generative AI, is transforming teaching, learning, and assessment processes in schools and colleges.
  • Opportunities such as personalization and support are offset by challenges in assessment, skill development, copyright, and data protection.
  • Empirical and conceptual foundations for responsible use have been limited thus far.

Potential Research Areas

  • Opportunities and risks of generative AI regarding diversity and inclusion
  • Impacts on assessment formats, evaluation, and academic integrity
  • Acceptance, use by faculty and students, AI literacy
  • Governance, guidelines, and data protection

Possible Research Methods

  • Conceptual (systematic literature review)
  • Empirical (interviews, surveys, experiments)
  • Technical/design-oriented (development approaches, Design Science Research)

Introductory Literature

Contact: Rizana Joers, rizana.joers[at]uni-kassel[dot]de

Motivation and Relevance

  • Compassion means that we not only recognize our own suffering or that of others, but also want to take proactive steps to address it.
  • Within the framework of Compassionate Technologies (which also includes Compassionate AI), technologies are developed, designed, and studied that proactively focus on reducing suffering and harm of all kinds.
  • Examples of potential suffering or harm include: addictions (brain rot, digital detox), fears about the future, overconsumption, inequalities, environmental damage, etc. A more comprehensive overview of different types of suffering can be found here.
  • The focus is on better addressing human needs and desires, improving how we interact with everyday technologies (e.g., TikTok, YouTube, cell phone use), and incorporating diverse cultures and contexts (e.g., specialized apps for specific illnesses).

 

Possible Areas of Research

  • Analysis of the behavior of existing (AI) technologies and suggestions for improvement; different target groups (children, adolescents, students, seniors, etc.); alternative approaches
  • Challenges and potential of AI companions, mitigation of existing risks
  • Ideas for technologies that promote empathy toward (1) others, (2) oneself, or (3) between people
  • How can technologies or AI, for example, support the improvement of well-being or the promotion of inclusivity?
  • What contribution can they make to various current challenges such as sensory overload, sustainability, etc.?

 

Possible research methods

  • Conceptual studies (systematic literature review)
  • Empirical studies (interviews, surveys, experiments)
  • Technical studies (development approaches, design science research)

 

Introductory Literature

  • Ciriello, R. F., Chen, A. Y., & Rubinsztein, Z. A. (2025). Compassionate AI Design, Governance, and Use. IEEE Transactions on Technology and Society, 6(3), 270–275. doi.org/10.1109/TTS.2025.3538125
  • Lee, M., Ackermans, S., van As, N., Chang, H., Lucas, E., & IJsselsteijn, W. (2019). Caring for Vincent: A Chatbot for Self-Compassion. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, CHI ’19, 1–13. doi.org/10.1145/3290605.3300932

Contact: Prof. Dr. Sebastian Lins, sebastian.lins[at]uni-kassel[dot]de

Motivation and Relevance

  • AI tools have long since become part of everyday work: Employees use language models for research, writing, coding, and decision-making. In response to the risks posed by AI, calls for a “human-in-the-loop” approach are growing louder: Humans are expected to review and interpret AI results and catch errors.
  • This is precisely where a contradiction arises: Studies on information security have shown for years that employees themselves do not behave securely enough—from phishing and passwords to the handling of sensitive data.
  • But those who fail to recognize their own security risks can hardly judge whether an AI is disclosing confidential information, making unsafe recommendations, or has been manipulated. The human as a control mechanism thus becomes a vulnerability rather than a safeguard.
  • In this context, new testing procedures, competency and training concepts, and other technical and organizational approaches are needed to effectively support people in monitoring AI and to enhance information security.

Possible Areas of Research

  • Secure Use of AI and Agents
  • The Tension Between “Human-in-the-Loop” and Ignorance & Overwhelm in Information Security

Possible research methods

  • Conceptual studies (systematic literature review)
  • Empirical studies (interviews, surveys, experiments)

Introductory Literature

Contact: Prof. Dr. Sebastian Lins, sebastian.lins[at]uni-kassel[dot]de

Motivation and Relevance

  • The shift of AI to the cloud as a service (AI as a Service) makes powerful models available to organizations that could never run them on their own. This makes AI more accessible, yet companies often struggle to utilize cloud-based AI services.
  • At the same time, computing power is no longer a question of “cloud or on-premises,” but rather one of distribution. An intermediate layer has emerged between central cloud data centers and end devices: fog nodes that process data where it is generated. Individual operating models thus merge into a seamless continuum in which workloads can be flexibly deployed.

Potential Areas of Research

  • What challenges do companies face when using cloud-based AI?
  • How can companies make effective use of the new technology known as fog computing?

Possible research methods

  • Empirical studies (interviews or surveys)

Introductory Literature