Wearables such as smartwatches provide an effective and efficient solution for continuous health data collection. In recent years, the number of people using these devices has steadily increased, along with advancements in their functionality, such as the ability to measure blood pressure, oxygen saturation, and ECG waves. This has led to an increase in opportunities for researchers to integrate these sensors into their research projects. However, collecting research data from consumer devices remains cumbersome and not easily applicable in practice. Addressing these challenges could help mitigate data shortages across research fields. Therefore, the aim of this systematic literature review is to explore and identify approaches and possibilities for using consumer wearables in scientific research.

Literature

• Dobson, R., Stowell, M., Warren, J., Tane, T., Ni, L., Gu, Y., McCool, J. & Whittaker, R. (2023). Use of consumer wearables in health research: issues and considerations. Journal of medical Internet research, 25, e52444.
• Shah, M. U., Rehman, U., Iqbal, F., Wahid, F., Hussain, M., & Arsalan, A. (2020, November). Access permissions for Apple Watch applications: A study on users' perceptions. In 2020 International Conference on Communications, Computing, Cybersecurity, and Informatics (CCCI) (pp. 1-7). IEEE.

Feature engineering is one of the most important steps to train accurate and precise machine learning algorithms. A traditional approach to derive features from the original dataset is to extract information and properties using manual calculation of statistical measures or definition of specific knowledge about the phenomenon, commonly referred to as hand-crafted features. In the field of health research, cardiac features are of particular importance because they capture essential physiological information. Among them, heart rate variability (HRV) is frequently used, since HRV features provide a relevant indicator of stress and other mental states. In recent years, a variety of approaches have been proposed to extract HRV features for this purpose. The aim of this systematic literature review is to identify hand-crafted features derived from HRV data that are used in the assessment of mental states. 

Literature

• Kim, H. G., Cheon, E. J., Bai, D. S., Lee, Y. H., & Koo, B. H. (2018). Stress and heart rate variability: a meta-analysis and review of the literature. Psychiatry investigation, 15(3), 235.
• Ishaque, S., Khan, N., & Krishnan, S. (2021). Trends in heart-rate variability signal analysis. Frontiers in Digital Health, 3, 639444.
• Sánchez, R. V., Macancela, J. C., Ortega, L. R., Cabrera, D., García Márquez, F. P., & Cerrada, M. (2024). Evaluation of hand-crafted feature extraction for fault diagnosis in rotating machinery: a survey. Sensors, 24(16), 5400.
• Singh, A. K., & Krishnan, S. (2023). ECG signal feature extraction trends in methods and applications. BioMedical Engineering OnLine, 22(1), 22.

With the growing popularity of wearables such as smartwatches or smart rings, physiological data can be easily tracked. Despite the known potentials for users, such as medical prevention or fitness tracking, personal physiological data can also be misused. When combined with non-physiological data, the predictive capacity of an algorithm could expand to surveillance of a user's cognitive and affective states. For instance, an employer could track the level of engagement and enthusiasm of employees while at work. With the increasing risk of using wearables far beyond their original purpose, the need for regulations and privacy concepts becomes more urgent. The aim of this systematic literature review is to explore the current state-of-the-art on privacy protection for physiological data measurements.

Literature

• Davis Iii, K. M., & Ruotsalo, T. (2025). Physiological data: Challenges for privacy and ethics. Computer, 58(1), 33-44.
• Tu, X., Niu, Z., Yin, J., Zhang, Y., Yang, M., Wei, L., Wang, Y., Fan, Z. & Zhao, J. (2026). Phys-EdiGAN: A privacy-preserving method for editing physiological signals in facial videos. Pattern Recognition, 169, 111966.

The spotlight in the discussion on how XAI can aid in achieving medical goals usually shines on the doctor who has to make complex, high-stakes decisions. However, the treatment decision does not depend solely on the doctor’s professional opinion. What must also be considered is the patients’ role in the decision to be made. Without their approval, the doctor cannot begin treatment, even if they are absolutely certain of its benefit to the patient. As AI methods see increased performance and applicability for medical use cases, the question of patient trust and understanding needs to be considered next to the perspective of medical professionals such as doctors or nurses. 

Therefore, in the scope of this seminar paper, a literature review will be conducted to explore how medical patients respond to explanations generated through XAI methods. 

Literature

• Lötsch, J., Kringel, D., & Ultsch, A. (2021). Explainable artificial intelligence (XAI) in biomedicine: Making AI decisions trustworthy for physicians and patients. BioMedInformatics, 2(1), 1-17.
• Weber, P., Heigl, R. M., & Baum, L. (2024). Understanding (X) AI in Healthcare: Patients’ Perspective on Global and Local Explanations. PACIS 2024 Proceedings. 10.

Die zunehmende Bedeutung der Circular Economy erfordert innovative digitale Lösungen, um nachhaltige Konsum- und Produktionsmuster zu fördern. Recommender Systeme – bekannt aus Online-Plattformen wie E-Commerce oder Streaming – können in diesem Kontext gezielt eingesetzt werden, um zirkuläre Geschäftsmodelle zu unterstützen und Ressourcenschonung zu fördern. Während klassische Empfehlungssysteme vor allem auf Umsatzsteigerung und Nutzerbindung abzielen, liegt der Fokus zirkulärer Recommender Systeme auf der Verlängerung von Produktlebenszyklen, der Wiederverwendung von Materialien sowie der Unterstützung nachhaltiger Kauf- und Nutzungsentscheidungen.

In der Seminararbeit wird untersucht, wie Algorithmen zur Personalisierung auf Nachhaltigkeitsziele ausgerichtet werden können. Themen sind unter anderem: Welche Datenquellen eignen sich für Empfehlungen im Bereich Sharing, Reuse oder Refurbishment? Wie lassen sich Präferenzen der Nutzer mit ökologischen Kriterien verbinden? Welche Rolle spielen Transparenz und Erklärbarkeit, um Vertrauen in Empfehlungen für nachhaltige Optionen zu schaffen?

Ziel der Arbeit ist es, Potenziale und Herausforderungen zirkulärer Recommender Systeme zu diskutieren und konkrete Anwendungsfelder – von Second-Hand-Plattformen über Produkt-Service-Systeme bis hin zu industriellen Materialkreisläufen – zu beleuchten. Die Teilnehmenden gewinnen ein tieferes Verständnis dafür, wie datengetriebene Personalisierung zu einer ressourcenschonenden und zukunftsfähigen Wirtschaft beitragen kann.

Literature

• Lu, J., Wu, D., Mao, M., Wang, W., & Zhang, G. (2015). Recommender system application developments: a survey. Decision Support Systems, 74, 12-32.
• Resnick, P., & Varian, H. R. (1997). Recommender systems. Communications of the ACM, 40(3), 56-58.
• van Capelleveen, G., Amrit, C., Zijm, H., Yazan, D. M., & Abdi, A. (2021). Toward building recommender systems for the circular economy: Exploring the perils of the European Waste Catalogue. Journal of Environmental Management, 277, 111430.

Künstliche Intelligenz verändert zunehmend die Art und Weise, wie Software entwickelt wird. Gerade im Bereich der Open-Source-Software (OSS) eröffnen KI-gestützte Werkzeuge neue Möglichkeiten: Von automatischer Code-Generierung über Fehlererkennung bis hin zur Dokumentation. Gleichzeitig entstehen jedoch auch neue Risiken, die für die Nachhaltigkeit und Sicherheit von Open-Source-Ökosystemen von großer Bedeutung sind.

Ein zentrales Problem ist die Qualität und Verlässlichkeit von durch KI erzeugtem Code. Automatisch generierte Lösungen können Sicherheitslücken, ineffiziente Strukturen oder rechtlich problematische Komponenten enthalten. Hinzu kommt die Gefahr der „Code Inflation“: Eine Flut an automatisch erstelltem, aber schwer wartbarem Code könnte die Übersichtlichkeit und Stabilität von Projekten gefährden. Auch ethische und rechtliche Fragen stellen sich – etwa in Bezug auf Urheberrechte, Lizenzkonflikte oder die Verantwortung für Fehler im KI-generierten Code.

Im Seminar werden diese Gefahren systematisch beleuchtet: Welche technischen Risiken bestehen durch fehlerhafte Generierung und unzureichende Validierung? Wie wirken sich KI-Tools auf die Governance und Community-Strukturen von Open-Source-Projekten aus? Und welche Maßnahmen sind denkbar, um Risiken zu minimieren – beispielsweise durch Richtlinien, Prüfmechanismen oder neue Formen der Zusammenarbeit?

Ziel ist es, Chancen und Gefahren der KI-Nutzung in der Open-Source-Softwareentwicklung kritisch zu reflektieren und Handlungsoptionen für eine verantwortungsbewusste Anwendung aufzuzeigen.

Literature

• Lin, R., Fu, Y., Yi, W., Yang, J., Cao, J., Dong, Z., ... & Li, H. (2024). Vulnerabilities and security patches detection in OSS: a survey. ACM Computing Surveys, 57(1), 1-37.
• Song, F., Agarwal, A., & Wen, W. (2024). The impact of generative AI on collaborative open-source software development: Evidence from GitHub Copilot. arXiv preprint arXiv:2410.02091.

Der Einstieg in Open-Source-Softwareprojekte ist für neue Beitragende oft mit erheblichen Hürden verbunden. Komplexe Codebasen, fragmentierte Dokumentation und unausgesprochene Community-Regeln können das Onboarding erschweren und zur Abbruchquote neuer Mitwirkender beitragen. Künstliche Intelligenz bietet hier innovative Ansätze, um den Prozess zu vereinfachen und Neulingen eine gezielte Unterstützung zu bieten.

KI-gestützte Mentoren können als intelligente Begleiter fungieren, die Fragen beantworten, technische Hintergründe erläutern oder den passenden Einstiegspunkt im Code aufzeigen. Durch Natural Language Processing und personalisierte Empfehlungen lassen sich Lernpfade an die Vorerfahrungen und Interessen der Beitragenden anpassen. Auch soziale Aspekte des Onboardings – etwa das Verstehen von Kommunikationsnormen oder die Integration in eine Community – können durch KI-gestützte Assistenzsysteme unterstützt werden.

Im Seminar wird untersucht, wie AI Mentoren technisch umgesetzt werden können, welche Datenquellen und Modelle dabei eine Rolle spielen und welche Chancen und Risiken mit ihrem Einsatz verbunden sind. Diskutiert werden auch Fragen der Akzeptanz: Inwiefern können KI-Mentoren menschliche Betreuung ergänzen, ohne sie zu ersetzen?

Ziel ist es, Potenziale und Grenzen KI-gestützter Onboarding-Ansätze zu beleuchten und ihre Bedeutung für die Nachhaltigkeit und Diversität von Open-Source-Communities herauszuarbeiten.

Literature

• Tan, X., Long, X., Zhu, Y., Shi, L., Lian, X., & Zhang, L. (2025). Revolutionizing Newcomers’ Onboarding Process in OSS Communities: The Future AI Mentor. Proceedings of the ACM on Software Engineering, 2(FSE), 1091-1113.

The evolution of digital platforms has revolutionized the future of work, particularly in the technology sector, where both the dynamics of hiring and the outsourcing of temporary work are under increasing scrutiny (Höllig et al., 2024). With the upcoming of the gig economy – a labour market characterized by the prevalence of short-term contracts or freelance work as opposed to permanent jobs – (De Stefano, 2015; e.g., Upwork, Freelancer.com, Clickworker, Crowdflower, 99designs, Jovoto, etc.), traditional labour practices are profoundly reshaped. Following the conceptualization by Schmidt (2017, p. 6), the focus of this seminar lays on digital services categorized under the term ‘cloud work’ (web-based). Beyond matching supply and demand, platform owners act as keystone orchestrators who design governance rules, allocate visibility and access, and control data infrastructures that shape who captures value and how. The seminar adopts a structural perspective on gig ecosystems – explicitly identifying and modeling roles/meta-roles, value-creation activities, and value flows among participants – by transferring to gig work the extended e3-value approach recently demonstrated for e-commerce ecosystems by Wulfert et al. (2024). That study integrated a systematic literature review with a multi-case analysis, introduced ecosystem segments into e3-value for higher-order abstraction, and derived five categories of value flows. In this seminar, the goal is to mirror this procedure in order to surface gig-work specific orchestration mechanisms (e.g., algorithmic management, escrow, dispute resolution, rating regimes) and their implications. Possible research questions are:

1. Which roles/meta-roles constitute value creation in gig ecosystems?
2. What core and supporting activities are performed by each role/meta-role?
3. Which value objects and value flows instantiate gig-work?

As a result, the seminar should provide an empirically grounded generic e3-value model (e.g., Gordihn et al., 2000; Pousttchi, 2008) of the gig-work ecosystem. The work by Wulfert et al. (2024) should act as methodological guidance for this seminar. 

Literature

• De Stefano, V. (2015). The rise of the just-in-time workforce: On-demand work, crowdwork, and labor protection in the gig-economy. Comp. Lab. L. & Pol'y J., 37, 471.
• Gordijn, J., Akkermans, H., & Van Vliet, H. (2000). What’s in an electronic business model? In R. Dieng & O. Corby (Eds.), EKAW 2020: Knowledge engineering and knowledge management methods, models, and tools (pp. 257–273). Springer. doi.org/10.1007/3-540-39967-4_19.
• Höllig, C. E., Tumasjan, A., & Lievens, F. (2024). What drives employers’ favorability ratings on employer review platforms? The role of symbolic, personal, and emotional content. International Journal of Selection and Assessment, 32(4), 579-593.
• Pousttchi, K. (2008). A modeling approach and reference models for the analysis of mobile payment use cases. Electronic Commerce Research and Applications, 7(2), 182–201. doi.org/10.1016/j.elerap.2007.07.001.
• Schmidt, F. A. (2017). Digital labour markets in the platform economy. Mapping the Political Challenges of Crowd Work and Gig Work, 7, 2016. Abgerufen am 01.09.2025 unter: library.fes.de/pdf-files/wiso/13164.pdf
• Wulfert, T., Woroch, R., & Strobel, G. (2024). Follow the flow: An exploratory multi-case study of value creation in e-commerce ecosystems. Information & Management, 61(8), 104035.

Low-code development continues to be a trend in software development. The term “low code” is used differently by vendors of low-code platforms and may, among other, refer to workflowmanagement systems, small database management systems, or application-development environments (Bock and Frank 2020). What is typically referred to as “low-code development” may involve a number of different techniques, all of which are aimed at increasing the productivity of development and enabling end users to develop their own applications (“enduser development” in this context is often referred to as “citizen development”). For example, low-code platforms may offer reference solutions so that developers do not need to develop new software from scratch every time. They may also offer visual development aids, such as a drag-and-drop GUI editor. However, most low-code vendors seem to be based almost exclusively on general-purpose languages such as Java or BPMN and make almost no use of domain-specific languages (Frank et al. 2021).

Domain-specific languages have been proposed as a more productive alternative to general-purpose languages at least since 1975 (Hammer 1975). They offer language constructs which embody relevant domain knowledge already, prohibiting any faulty implementations that would have been possible with general-purpose languages. Benefits of domain-specific languages over general-purpose languages have been known for long, but the use of DSLs is also faced with several challenges (Kelly and Tolvanen 2008; Kärnä et al 2009; Haugen 2013).

In this seminar paper, you are asked to analyze the prospects and challenges of using domain-specific languages in low-code development. Based on a literature review of potential benefits and challenges, you are asked to analyze two low-code platforms (which platforms should be discussed with the advisor) with regards to how potential benefits of DSLs can be realized and which challenges users are faced with.

Literature

• Bock AC, Frank U (2021) Low-Code Platform. Business and Information Systems Engineering 63(6):733–740
• Frank U, Maier P, Bock A (2021) Low Code Platforms: Promises, Concepts and Prospects. A Comparative Study of Ten Systems. ICB Research Report, vol 70
• Hammer M (1975) The Design of Usable Programming Languages. ACM '75: Proceedings of the 1975 annual conference, pp 225–229
• Haugen Ø (2013) Domain-Specific Languages and Standardization: Friends or Foes? In: ReinhartzBerger I et al (eds). Domain Engineering: Product Lines, Languages, and Conceptual Models. Springer: Berlin, Heidelberg, pp 159–186
• Kärnä J, Tolvanen J-P, Kelly S (2009) Evaluating the Use of Domain-Specific Modeling in Practice. DSM '09: Proceedings of the 9th OOPSLA Workshop on Domain-Specific Modeling, pp 14–20
• Kelly S, Tolvanen J-P (2008) Domain-Specific Modeling: Enabling Full Code Generation. John Wiley & Sons: Hoboken, NJ

Although organizations are increasingly permeated by information technology (IT) systems, questions concerning their actual profitability often remain unresolved. The so-called concept of IS business value and methods for its assessment are still subject to debate and lack a universally accepted approach (Melville et al., 2004; Schryen, 2013). Since the monetary benefits of IT are directly linked to the specific functions and organizational conditions of a given IT system, it appears reasonable to develop reference effect catalogs which can reference a specific IT system or a certain type of organization. Such reference catalogs list potential impacts of IT systems and thereby enable an analysis of whether these effects manifest within a given organizational context (Schütte et al., 2022; Seufert et al., 2021). However, the creation and continuous maintenance of such catalogs is a demanding task, often requiring specialized domain knowledge possessed primarily by experts (Schütte et al., 2022).

The recent emergence of large language models (LLMs), such as ChatGPT, suggests the potential to produce results that are comparable to, or in some cases exceed, those generated by human experts. This development introduces the possibility of creating and maintaining reference effect catalogs with the support of LLMs. Nevertheless, the concrete procedures for employing LLMs in this context, as well as the question of whether they actually outperform existing references catalogs (cf. Schubert & Williams, 2009; Shang & Seddon, 2002), remain unclear. Consequently, a systematic analysis and evaluation of how LLMs can support the construction and application of reference effect catalogs is required. The topic therefore aims at systematically experimenting what options LLMs offer in the creation of reference catalogs and evaluating of how they could perform against traditional reference catalogs. 

Literature

• Melville, Kraemer, & Gurbaxani. (2004). Review: Information Technology and Organizational Performance: An Integrative Model of IT Business Value. MIS Quarterly, 28(2), 283. doi.org/10.2307/25148636
• Schryen, G. (2013). Revisiting IS business value research: what we already know, what we still need to know, and how we can get there. European Journal of Information Systems, 22(2), 139–169. doi.org/10.1057/ejis.2012.45
• Schubert, P., & Williams, S. P. (2009). An Extended Framework for Comparing Expectations and An Extended Framework for Comparing Expectations and Realized Benefits of Enterprise Systems Implementations. AMCIS 2009 Proceedings.
• Schütte, R., Seufert, S., & Wulfert, T. (2022). IT-Systeme wirtschaftlich verstehen und gestalten. Springer Fachmedien Wiesbaden GmbH.
• Seufert, S., Wulfert, T., & Wernsdörfer, J. E. (2021). Towards a reference value catalogue for a company-specific assessment of the IT business value -proposing a taxonomy to select IT impacts from existing catalogues. ECIS 2021 Research-in-Progress Papers 34.
• Shang, S., & Seddon, P. B. (2002). Assessing and managing the benefits of enterprise systems: the business manager’s perspective. Information Systems Journal, 12(4), 271–299. doi.org/10.1046/j.1365-2575.2002.00132.x

Die Einführung von KI-Technologien am Arbeitsplatz stellt nicht nur eine technische, sondern vor allem eine kulturelle Herausforderung dar. Viele Mitarbeiter fürchten um ihre Rolle oder zögern, KI-Systeme aktiv zu nutzen, was zu Widerstand und geringerer Kooperationsbereitschaft führen kann. „Psychologische Sicherheit“ (Edmondson, 1999) – das Vertrauen darauf, ohne Angst vor negativen Konsequenzen Fragen zu stellen, Fehler einzugestehen oder neue Ideen einzubringen – gilt als Schlüsselfaktor für eine erfolgreiche Mensch-KI-Interaktion. Fehlt dieses Sicherheitsgefühl, validieren Mitarbeiter KI-Ergebnisse nicht kritisch oder vermeiden gar den Einsatz neuer KI-Tools, wodurch sowohl Lernprozesse als auch Innovationspotenzial gehemmt werden. Bisher fehlt jedoch eine detaillierte Untersuchung, wie psychologische Sicherheit die Akzeptanz von KI beeinflusst und durch welche Maßnahmen Führungskräfte ein solches Vertrauensklima gezielt fördern können. Mögliche Forschungsfragen sind:

1. Welche Bedeutung hat psychologische Sicherheit für die erfolgreiche Mensch-KI-Kollaboration im Arbeitskontext?
2. Welche Faktoren fördern bzw. hemmen psychologische Sicherheit bei der Einführung von KI?

Die vorliegende Seminararbeit zielt darauf ab, basierend auf Experteninterviews zentrale Erfolgsfaktoren herauszuarbeiten und erste Implikationen für die Gestaltung einer vertrauensfördernden Mensch-KI-Kollaboration abzuleiten.

Literature

• Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383.
• Kim, B. J., Kim, M. J., & Lee, J. (2025). The dark side of artificial intelligence adoption: linking artificial intelligence adoption to employee depression via psychological safety and ethical leadership. Humanities and Social Sciences Communications, 12(1), 1-14.
• Raisch, S., & Krakowski, S. (2021). Artificial intelligence and management: The automation-augmentation paradox. Academy of Management Review, 46(1), 192–210

Die rasche Verbreitung von KI in Unternehmen verändert Arbeitsprozesse, Rollenprofile und Kompetenzanforderungen grundlegend. Routinetätigkeiten werden automatisiert, während für Mitarbeiter vermehrt strategische, analytische und überwachende Aufgaben entstehen. Dadurch verschiebt sich der Kompetenzbedarf: Datenanalyse, digitale Kompetenzen, komplexe kognitive Fähigkeiten, Entscheidungsfähigkeit und kontinuierliches Lernen (Jaiswal, Arun & Varma, 2023) werden zu unerlässlichen Fähigkeiten quer durch alle Funktionen. Viele Organisationen laufen Gefahr, von der KI-Entwicklung überholt zu werden, da ihre Workforce-Strategie (Weiterbildung, Umschulung) nicht Schritt hält. Erste Untersuchungen zeigen einen deutlichen Skill-Gap: Mitarbeiter fühlen sich im Umgang mit KI oft unzureichend geschult, was die Ausschöpfung des vollen Potenzials von KI verhindert. Zudem besteht das Risiko, dass „Cognitive Offloading“ an KI-Systeme langfristig zu Kompetenzverlust führt, wenn Unternehmen keine Gegenstrategien zum Erhalt und Ausbau menschlicher Fähigkeiten entwickeln. Es stellt sich somit die Frage, wie Unternehmen ihre Personalentwicklung an die KI-Transformation anpassen müssen, um sowohl den Verlust von Fachwissen zu verhindern als auch neue Kompetenzen effektiv aufzubauen. Mögliche Forschungsfragen sind:

1. Welche neuen Kompetenzanforderungen ergeben sich durch den Einsatz von KI in Organisationen?
2. Welche Strategien zum Upskilling und Reskilling wenden Unternehmen aktuell an?
3. Mit welchen Herausforderungen und Erfolgsfaktoren sind diese Strategien verbunden?

Die Arbeit verfolgt das Ziel, anhand qualitativer Interviews zu untersuchen, welche Ansätze Unternehmen im Umgang mit neuen Kompetenzanforderungen durch KI entwickeln und welche Herausforderungen dabei auftreten. Dabei sollen einerseits zentrale neue Kompetenzfelder (z. B. Datenkompetenz, kritisches Denken, Zusammenarbeit mit KI-Systemen) identifiziert und andererseits die praktischen Strategien der Qualifikationsentwicklung (z. B. interne Trainings, Lernplattformen, Coaching-Programme) vergleichend herausgearbeitet werden. Auf dieser Basis sollen Handlungsempfehlungen formuliert werden, wie Organisationen Qualifikationsentwicklung im Kontext von KI erfolgreich gestalten können.

Literature

• Jaiswal, A., Arun, C. J., & Varma, A. (2023). Rebooting employees: Upskilling for artificial intelligence in multinational corporations. In Artificial intelligence and international HRM (pp. 114-143). Routledge. - Morandini, S., Fraboni, F., De Angelis, M., Puzzo, G., Giusino, D., & Pietrantoni, L. (2023). The impact of artificial intelligence on workers’ skills: Upskilling and reskilling in organisations. Informing Science, 26, 39-68.
• Uren, V., & Edwards, J. S. (2023). Technology readiness and the organizational journey towards AI adoption: An empirical study. International Journal of Information Management, 68, 102588.

Conversational AI agents – such as chatbots and voice assistants – have emerged as a transformative technology in the retail sector, offering new types of interaction between retailers and customers (Arce-Urriza et al., 2021). These systems can simulate natural language conversations and are increasingly deployed to assist with different tasks along a customer’s journey, such as product discovery or customer service. At the same time, they promise a fast return on investment and relieve human employees of repetitive tasks, allowing them to focus on higher-value customer engagements (Chong et al., 2021). Nevertheless, many consumers remain skeptical about the capabilities of conversational AI agents (Chong et al., 2021).

This seminar paper involves a structured literature review to systematically identify and analyze the potential adoption cases of conversational AI agents in the retail sector. The review will be guided by the Technology Acceptance Model (TAM), focusing on the constructs of perceived usefulness and perceived ease of use to assess the customer acceptance (Davis, 1989). Therefore, the aim of this seminar paper is to deliver a synthesis of adoption scenarios, highlight how different design features (e.g., personalization) of conversational AI agents influence customer acceptance in retail, and assess the applicability and limitations of TAM in this context.

Literature

• Arce-Urriza, M., Chocarro, R., Cortiñas, M., & Marcos-Matás, G. (2025). From familiarity to acceptance: The impact of Generative Artificial Intelligence on consumer adoption of retail chatbots. Journal of Retailing and Consumer Services, 84, 104234. doi.org/10.1016/j.jretconser.2025.104234
• Banala, T. V., Pasupuleti, R. S., Thiyyagura, D., & H, D. P. (2024). Measuring the Role of Trust and Information Quality on Intention to Use AI Chatbots in Mobile Shopping: A Structured Equation Modeling Approach (pp. 284–292). doi.org/10.1109/icdcc62744.2024.10961076
• Chong, T., Yu, T., Keeling, D. I., & De Ruyter, K. (2021). AI-chatbots on the services frontline addressing the challenges and opportunities of agency. Journal of Retailing and Consumer Services, 63, 102735. doi.org/10.1016/j.jretconser.2021.102735
• Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of Information Technology. MIS Quarterly, 13(3), 319. doi.org/10.2307/249008 ▪ Manzo, D. S. H., Jiang, Y., Elyan, E., & Isaacs, J. (2024). Artificial Intelligence-Based Conversational Agents Used For Sustainable Fashion: Systematic Literature Review. International Journal of Human-Computer Interaction, 1–13. doi.org/10.1080/10447318.2024.2352920

Artificial Intelligence (AI) offers new possibilities for conducting benchmarking in Information Systems (IS). Traditional benchmarking methods—based on static datasets, manual metrics selection, and episodic comparisons—are increasingly supported or transformed by AI-driven approaches. By enabling automatic data collection, real-time analysis, predictive comparisons, and adaptive benchmarking frameworks, AI can enhance the efficiency, scalability, and relevance of benchmarking practices at the levels of processes, organizations, capabilities, and strategies.

However, the integration of AI also raises open questions. How can AI tools ensure transparency and validity in benchmarking? What risks arise from biases, over-automation, or misalignment with organizational goals? To what extent can AI enable continuous, context-aware benchmarking in volatile and complex environments?

The student’s task is to develop a perspective on how AI may transform benchmarking in IS. The research will (1) outline traditional benchmarking approaches, (2) identify potential ways in which AI techniques (e.g., machine learning, natural language processing, anomaly detection) could support or reshape these approaches, and (3) critically reflect on opportunities, challenges, and risks. As the literature on “benchmarking with AI” is still emerging, students are expected to combine insights from related fields with their own structured argumentation to propose possible future directions.

Maturity models provide structured frameworks for assessing how organizations develop their processes, capabilities, and governance practices over time. Traditionally, these assessments rely on expert-driven evaluations, static surveys, and periodic reviews. Artificial Intelligence (AI) opens up new possibilities for supporting maturity assessments by enabling automated data collection, real-time monitoring, predictive analytics, and adaptive pathways that respond to changing organizational contexts.

At the same time, the integration of AI raises important questions: How can AI improve the validity and scalability of maturity assessments? What risks emerge from over-automation, biased data, or lack of transparency? And to what extent can AI help organizations make maturity assessments more continuous, dynamic, and context-sensitive?

The student’s task is to develop a perspective on how AI may transform maturity assessments in IS. The research will (1) outline the foundations of maturity models and their traditional limitations, (2) identify potential roles of AI techniques (e.g., machine learning, natural language processing, anomaly detection) in supporting or reshaping these assessments, and (3) critically reflect on opportunities, risks, and design principles for next-generation maturity models. Since the literature on “maturity assessments with AI” is still emerging, students are expected to combine theoretical insights with their own structured reasoning to propose possible future directions.

Reference models in Information Systems (IS) traditionally serve as standardized abstractions capturing common structures, processes, or governance practices across domains. Artificial Intelligence (AI) opens up new possibilities for supporting both the creation and the use of such models. On the creation side, AI could help identify patterns across large datasets, extract reusable structures, or automate parts of the modeling process. On the usage side, AI might enable real-time instantiation, adaptive customization, or dynamic alignment of models with specific organizational contexts.

The student’s task is to develop a perspective on how AI may transform reference modeling in IS. The research will (1) review the foundations of reference modeling and its methodological approaches, (2) explore potential applications of AI techniques (e.g., machine learning, natural language processing, generative approaches) for creating and applying reference models, and (3) critically reflect on the opportunities, risks, and implications of AI-supported reference modeling. Since the literature on “reference modeling with AI” is still very limited, students are expected to build their own structured argumentation by combining concepts from both fields and proposing possible future directions.

Information Systems (IS) research and practice often employ benchmarking, maturity assessments, and reference modeling as separate approaches for evaluation, improvement, and standardization. Yet, these concepts share important similarities: benchmarking provides external comparators and performance standards, maturity assessments track internal capability evolution, and reference modeling offers reusable structures for processes, organizations, and governance. Exploring their potential connections raises the question of whether they can be combined into more holistic frameworks for organizational design and IS development.

The student’s task is to develop a perspective on how benchmarking, maturity assessments, and reference modeling could be linked in IS. The research will (1) define and differentiate these concepts clearly, (2) explore possible intersections and complementarities between them, and (3) propose a conceptual model or framework showing how an organization might integrate all three to create a more comprehensive assessment and improvement cycle. Since explicit literature on their integration is limited, students are expected to synthesize ideas from each field and argue critically about opportunities, risks, and methodological implications of such an approach.

As Information Systems grow in complexity and organizational environments become more dynamic, effective benchmarking, maturity assessments, and reference modeling demand more than traditional technical know-how. Beyond established analytical methods, future organizational performance in these areas may depend on a combination of cognitive, analytical, and human skills—such as data literacy, systems thinking, adaptability, collaboration, and ethical awareness.

The student’s task is to develop a perspective on which skills will remain relevant—or become newly essential—for applying benchmarking, maturity assessments, and reference modeling in IS. The research will (1) review existing “future skills” and competency frameworks, (2) consider how these could be linked to the methodological needs of IS evaluation, and (3) propose a justified set of skills that individuals and teams should cultivate. Since direct literature on skills for these specific IS methods is limited, students are expected to synthesize insights from related fields and build their own structured argument for how organizations can prepare for future challenges.

Accurate assessment of one's own capabilities as well as those of the AI system is fundamental for effective human-AI collaboration. This "metaknowledge" refers to the ability to precisely evaluate both one's own capacities and those of collaborative partners, making informed decisions about task allocation and responsibilities based on this assessment. In human-AI teams, metaknowledge enables participants to decide when they should collaborate and how to interpret, accept, or challenge each other's outputs. This metaknowledge about cognitive processes – both one's own and the AI's – enables not only strategic task division but also continuous monitoring and adjustment of the collaborative workflow. While over- or underestimation of AI capabilities leads to suboptimal utilization, sufficient metaknowledge enables balanced task allocation between human and AI. However, the practical development and application of metaknowledge in human-AI teams presents ongoing challenges (e.g., effective governance) that warrant further investigation.

The objective of this thesis is to explore through semi-structured interviews with practitioners how metaknowledge develops in real-world human-AI collaboration contexts, investigating the challenges professionals face in accurately assessing AI capabilities, the strategies they employ to improve collaboration effectiveness, and the organizational factors that support or hinder metacognitive development in human-AI teams. Through qualitative analysis of practitioner experiences, this research aims to identify patterns in metaknowledge acquisition and provide insights for designing better collaborative agentic IS implementations.

Literature

• Colville, S., & Ostern, N. (2024): Trust and Distrust in GAI Applications: The Role of AI Literacy and Metaknowledge. ICIS 2024 Proceedings.
• Fügener, A., Grahl, J., Gupta, A., & Ketter, W. (2021): Cognitive Challenges in Human–Artificial Intelligence Collaboration: Investigating the Path Toward Productive Delegation. Information Systems Research, 33(2), 678–696. doi:10.1287/isre.2021.1079
• Han, Y., & Dunning, D. (2024): Metaknowledge of Experts Versus Nonexperts: Do Experts Know Better What They Do and Do Not Know? Journal of Behavioral Decision Making, 37(2), e2375. doi:10.1002/bdm.2375
• Jussupow, E., Spohrer, K., Heinzl, A., & Gawlitza, J. (2021): Augmenting Medical Diagnosis Decisions? An Investigation into Physicians’ Decision-Making Process with Artificial Intelligence. Information Systems Research, 32(3), 713–735. doi:10.1287/isre.2020.0980
• Strunk, J., Banh, L., Nissen, A., Strobel, G., & Smolnik, S. (2024): To Delegate or Not to Delegate? Factors Influencing Human-Agentic IS Interaction. ICIS 2024 Proceedings.

Die rasanten Fortschritte in der Generativen Künstlichen Intelligenz (GenAI) verändern nicht nur die Art und Weise, wie Inhalte erzeugt werden, sondern eröffnen zunehmend neue Möglichkeiten für autonome, zielgerichtete Entscheidungs- und Handlungsprozesse. Diese Entwicklung geht mit einer physischen oder digitalen Verkörperung von KI einher. Besonders die physische Verkörperung – etwa in Form von Drohnen, Rover, humanoiden oder tierähnlichen Robotern – ermöglicht eine neue Qualität der Kollaboration zwischen Mensch und Maschine sowie zwischen unterschiedlichen Maschinenformen, einschließlich rein digitaler KI Systeme.

Im Kontext dieser verkörperten Intelligenz kommt der Theorie der Schwarmintelligenz eine besondere Bedeutung zu. Sie beschreibt, wie sich komplexes und intelligentes Verhalten aus der Interaktion vieler relativ einfacher Einheiten ergibt – ganz ohne zentrale Steuerung. Aus der Biologie inspiriert, zeigt Schwarmintelligenz, wie durch Selbstorganisation, lokale Interaktion und Emergenz kollektive Lösungen entstehen können, die robust, flexibel und anpassungsfähig sind. In der KI-Forschung und Robotik wurde dieser Ansatz bereits in Optimierungsverfahren (z. B. Particle Swarm Optimization, Ant Colony Optimization) sowie in Roboterschwärmen erprobt. Gerade im Zusammenspiel mit generativer KI entstehen neue Fragen: Wie können lernende, kreative Systeme in schwarmartige Strukturen eingebunden werden? Welche neuen Formen kollektiver Intelligenz lassen sich dadurch erschließen?

Ziel der Seminararbeit ist es, auf Grundlage einer systematischen Literaturrecherche die bisherige Forschung zu Schwarmintelligenz im Zusammenspiel mit Generativer KI und Robotik zu erfassen und kritisch zu reflektieren. Im Zentrum stehen dabei drei Leitfragen:

• Welche Anwendungsszenarien ergeben sich für den Einsatz schwarmintelligenter, generativer KI-Systeme und welche Lösungen gibt es bereits? Was ist der aktuelle Stand der Forschung?
• Welche Vorteile und Herausforderungen zeigen sich in bisherigen Ansätzen?
• Wo bestehen Forschungslücken, und an welchen Stellen sollte zukünftige Forschung gezielt ansetzen?

Die Arbeit soll damit ein fundiertes Bild der Potenziale und Grenzen zeichnen, die sich aus der Verbindung von Schwarmintelligenz und Generativer KI ergeben – sowohl für physisch verkörperte Systeme als auch für digitale KI-Formen.

Literature

• Banh, L. & Strobel, G. (2023). Generative artificial intelligence. Electronic Markets 33(63). DOI: 10.1007/s12525-023-00680-1
• Duan, J., Yu, S., Tan, H. L., Zhu, H., & Tan, C. (2022). A Survey of Embodied AI: From Simulators to Research Tasks. IEEE Transactions on Emerging Topics in Computational Intelligence, 6(2), 230–244. doi.org/10.1109/TETCI.2022.3141105
• Feuerriegel S, Hartmann J, Janiesch C, Zschech P (2024). Generative AI. Business & Information Systems Engineering 66(1):111–126. doi:10.1007/s12599-023-00834-7
• Laufer, J., Banh, L., & Strobel, G. Bridging Mind and Matter: A Taxonomy of Embodied Generative AI.
• Pfeifer, R., & Iida, F. (2004). Embodied Artificial Intelligence: Trends and Challenges. In D. Hutchison, T. Kanade, J. Kittler, J. M. Kleinberg, F. Mattern, J. C. Mitchell, M. Naor, O. Nierstrasz, C. Pandu Rangan, B. Steffen, M. Sudan, D. Terzopoulos, D. Tygar, M. Y. Vardi, G. Weikum, F. Iida, R. Pfeifer, L. Steels, & Y. Kuniyoshi (Eds.), Lecture Notes in Computer Science. Embodied Artificial Intelligence (Vol. 3139, pp. 1–26). Springer Berlin Heidelberg. doi.org/10.1007/978-3-540-27833-7_1
• Paolo, G., Gonzalez-Billandon, J., & Kégl, B. (2024). Position: A Call for Embodied AI. In Forty-first International Conference on Machine Learning. (Vol. 235, pp. 39493–39508).

Themenspezifische Literatur

• Ahmed, H., & Glasgow, J. (2012). Swarm intelligence: concepts, models and applications. School Of Computing, Queens University Technical Report.
• Chakraborty, A., & Kar, A. K. (2017). Swarm intelligence: A review of algorithms. Nature-inspired computing and optimization, 475-494.
• Rosenberg, L., Schumann, H., Dishop, C., Willcox, G., Woolley, A., & Mani, G. (2025). Large-scale Group Brainstorming and Deliberation using Swarm Intelligence and Generative AI. In Proceedings of the 27th International Conference on Enterprise Information: ICEIS.
• Zhang, Z., Li, X., & Pan, A. (2024, May). Generative Intelligence-Based Swarm Robots Control and Human-Robot Symbiotic Society. In 2024 16th International Conference on Advanced Computational Intelligence (ICACI) (pp. 99-106). IEEE.

In recent years, digital platforms have emerged as key instruments for fostering value creation across a wide range of industries. Companies such as Apple, Amazon, and Google illustrate how digital platforms can offer significant competitive advantages by strategically steering value creation. The integration of complementary offerings from loosely coupled and independent actors is a critical success factor in platform-based ecosystems. Platform operators act as orchestrators who must align the interests of heterogeneous stakeholders. Against this backdrop, the focus of this thematic section is to review the current state of research in the field of platform governance.

Research Questions:

• RQ1: What are the central topics addressed by Information Systems (IS) research in the context of platforms, ecosystems, and governance?
• RQ2: Which aspects of platform governance should be explored in future research?

Methodology:
Systematic literature review of leading academic journals (Basket of Eight) and major conferences (ICIS, ECIS, PACIS, AMCIS, HICSS).

Objective:
The thesis project seeks to conduct a systematic review of the current state of research in the field of platform governance. The objective is to identify key research streams and present them within a structured academic framework. Based on the literature review, the thesis will synthesize the findings into a governance framework that reflects the most relevant theoretical and conceptual perspectives and can serve as a foundation for future academic inquiry. A central focus will be on the theoretical lenses through which governance mechanisms have been studied. Building on these insights, the candidate is expected to develop a research agenda that systematically outlines promising directions for future research in this domain.

Literature

• Hein, Andreas (2020): Digital Platform Ecosystems: Emergence and Value Co-Creation Mechanisms. Technischen Universität München, München.
• Jacobides, Michael G.; Cennamo, Carmelo; Gawer, Annabelle (2018): Towards a theory of ecosystems. In Strat. Mgmt. J. 39 (8), pp. 2255–2276. DOI: 10.1002/smj.2904.
• Tiwana, Amrit (2014): Platform Ecosystems: Elsevier, Kapitel 6
• Hein, Andreas; Schreieck, Maximilian; Wiesche, Manuel; Krcmar, Helmut (2016): Multiple-Case Analysis on Governance Mechanisms of Multi-Sided Platforms. In : Multikonferenz Wirtschaftsinformatik. Ilmenau, Germany.
• Staub, Nicola; Haki, Kazem; Aier, Stephan; Winter, Robert (2022): Governance Mechanisms in Digital Platform Ecosystems: Addressing the Generativity-Control Tension. In Communications of the Association for Information Systems 51 (1), pp. 906–939. DOI: 10.17705/1CAIS.05137.
• Bandara, Wasana; Furtmueller, Elfi; Gorbacheva, Elena; Miskon, Suraya; Beekhuyzen, Jenine (2015): Achieving Rigor in Literature Reviews: Insights from Qualitative Data Analysis and Tool-Support. In CAIS 37. DOI: 10.17705/1CAIS.03708.
• vom Brocke, Jan; Simons, Alexander; Niehaves, Björn; Riemer, Kai; Plattfaut, Ralf; Cleven, Anne (2009): Reconstructing the giant: On the importance of rigour in documenting the literature search process. In : 17th European Conference on Information Systems (ECIS 2009). Verona, Italy, pp. 2206–2217. Available online at aisel.aisnet.org/ecis2009/161.
• Webster, Jane; Watson, Richard T. (2002): Analyzing the Past to Prepare for the Future: Writing a Literature Review. In MIS Quarterly 26 (2), pp. xiii–xxiii.

The increasing integration of generative AI (GenAI) into knowledge work introduces new forms of collaboration between humans and intelligent systems. Beyond enhancing efficiency and productivity, these tools affect fundamental cognitive and physiological processes. For example, they may alter how individuals allocate attention, regulate stress, perceive workload, or maintain focus in complex tasks. Users often rely on GenAI to reduce mental strain by delegating demanding activities such as synthesizing information, generating creative content, or planning structured workflows. Cognitive offloading is one such phenomenon, but it represents only a part of the broader spectrum of changes GenAI introduces into human cognitive and physiological functioning. In light of increasing research that focuses on productivity gains, financial impacts, or the changing future of work, taking in a human-centered perspective to assess the (neuro-)physiological effects GenAI exerts on its users at the core remains essential to shape a sustainable and responsible usage of it.

Hence, this seminar paper should aim to investigate these individual implications at the intersection of Information Systems (IS) and NeuroIS research. By conducting a systematic literature review (SLR), the work will synthesize (1) which individual effects are associated within human–(Gen)AI collaboration (e.g., stress, mental load, flow), (2) how prior studies have conceptualized and measured these effects on a neuro-physiological level, and (3) which methods have been employed to measure these aspects (e.g., EEG, eye-tracking, heart-rate variability). The goal is to identify existing approaches, highlight research gaps, and outline promising directions for future work on the physiological dimensions of human–GenAI collaboration.

Literature

• Banh, L.; Stangl, F.J.; Strobel, G.; Riedl, R.: Exploring the NeuroIS Potential for Generative Artificial Intelligence: Findings from a Literature Review. In: Davis, F.D.; Riedl, R.; vom Brocke, J.; Léger, P.-M.; Randolph, A.B.; Müller-Putz, G.R. (Hrsg.): Information Systems and Neuroscience NeuroIS Retreat 2025, Vienna, Austria. Springer, Cham, 2025.
• Fügener, A., Grahl, J., Gupta, A., Ketter, W.: Cognitive Challenges in Human–Artificial Intelligence Collaboration: Investigating the Path Toward Productive Delegation. Information Systems Research, vol. 33, 678–696 (2022). doi:10.1287/isre.2021.1079
• Grinschgl, S., Neubauer, A.C.: Supporting Cognition With Modern Technology: Distributed Cognition Today and in an AI-Enhanced Future. Front. Artif. Intell.,vol. 5, 908261 (2022). doi: 10.3389/frai.2022.908261
• Kosmyna, N., Hauptmann, E., Yuan, Y. T., Situ, J., Liao, X.-H., Beresnitzky, A. V., Braunstein, I., & Maes, P. (2025). Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task. arXiv. https://doi.org/10.48550/ARXIV.2506.08872
• Riedl, R., Léger, P.-M.: Fundamentals of NeuroIS. Information Systems and theBrain. Springer Berlin Heidelberg, Berlin, Heidelberg (2016). ISBN: 978-3-662-45091-8
• Ritz, E., Freise, L.R., Li, M.M.: Offloading to Digital Minds: How Generative AI Can Help to Craft Jobs. In: Davis, F.D., Riedl, R., vom Brocke, J., Léger, P.-M., Randolph, A.B., Müller-Putz, G.R. (eds.) Information Systems and Neuroscience. NeuroIS Retreat 2024, Vienna, Austria. Lecture Notes in Information Systems and Organization (2024)
• Schulz, T., Knierim, M.T., Weinhardt, C.: How Generative AI-Assistance Impacts Cognitive Load During Knowledge Work: A Study Proposal. In: Davis,F.D., Riedl, R., vom Brocke, J., Léger, P.-M., Randolph, A.B., Müller-Putz, G.R. (eds.) Information Systems and Neuroscience. NeuroIS Retreat 2024, Vienna, Austria. Lecture Notes in Information Systems and Organization (2024)
• Tadson, B., Krisam, C., Mädche, A.: Understanding Overreliance in Human-GenAI Interactions and Its Impact on Divergent Thinking: A NeuroIS Study. In: Davis,F.D., Riedl, R., vom Brocke, J., Léger, P.-M., Randolph, A.B., Müller-Putz, G.R. (eds.) Information Systems and Neuroscience. NeuroIS Retreat 2025, Vienna, Austria. Lecture Notes in Information Systems and Organization (2025)

The Symbol Grounding Problem (SGP) questions how symbols in AI systems can be meaningfully connected to real-world entities. The philosophical version of SGP emerges from the Computational Theory of Mind (CTM) which suggests that for AI to truly "understand," its symbols must be grounded in reality. However, recent arguments, such as those by Reto Gubelmann (2024), argues against the common belief that large language models (LLMs) fall prey to the SGP, challenging the claims of Bender and Koller (2020), by proposing a pragmatist norm-based approach to meaning, which suggests that large language models (LLMs) do not need grounding in the classical sense to demonstrate understanding.

This seminar will critically examine:

• The Symbol Grounding Problem: What it is, why it matters, and how it applies to AI.
• Gubelmann’s Pragmatic Argument: Why he claims LLMs escape the SGP.
• Alternative Theories of Meaning: CTM, Representationalism, and Pragmatism.
• Empirical Evidence & Open Questions: What approaches can help resolve this debate?

Innovative Twist on the Philosophical Octopus Test

This topic examines a creative extension of the Octopus Test, initially proposed by Bender and Koller (2020). It focuses on the interaction of large language model agents within dynamic environments and explores whether LLMs can autonomously form and stabilize new pragmatic norms through mutual interactions, highlighting implications for autonomous AI community behaviors.

Potential Research Questions:

• Can large language models autonomously develop new, stable "pragmatic norms" when placed into interactive communities with other LLMs, and how would these AI-invented norms evolve over generations of agent interaction?
• Can a community of interacting LLM-driven agents develop a language that's pragmatically comprehensible to humans, despite containing words and structures without explicit grounding or corresponding real-world references?

Literature

• De Curtò, J., & De Zarzà, I. (2025). LLM-Driven Social Influence for Cooperative Behavior in Multi-Agent Systems. IEEE Access.
• Piao, J., Yan, Y., Zhang, J., Li, N., Yan, J., Lan, X., ... & Li, Y. (2025). AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society. arXiv preprint arXiv:2502.08691.
• Fang, S., Liu, J., Xu, C., Lv, C., Hang, P., & Sun, J. (2025). Interact, Instruct to Improve: A LLM-Driven Parallel Actor-Reasoner Framework for Enhancing Autonomous Vehicle Interactions. arXiv preprint arXiv:2503.00502.
• Förster, F., Saunders, J., & Nehaniv, C. L. (2017). Robots that say “no” Affective symbol grounding and the case of intent interpretations. IEEE Transactions on Cognitive and Developmental Systems, 10(3), 530-544.
• Wang, J. Y., Sukiennik, N., Li, T., Su, W., Hao, Q., Xu, J., ... & Li, Y. (2024). A Survey on Human-Centric LLMs. arXiv preprint arXiv:2411.14491.
• Emily Bender and Alexander Koller. 2020. Climbing towards nlu: On meaning, form, and understanding in the age of data. Proceedings ofthe 58th Annual Meeting ofthe Association for Computational Linguistics, pages 5185–5198.
• Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28(1), 3–71.
• Glüer, K., Wikforss, Å., & Ganapini, M. (2024). The Normativity of Meaning and Content. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (Fall 2024). Metaphysics Research Lab, Stanford University.
• Gubelmann, R. (2023). A Loosely Wittgensteinian Conception of the Linguistic Understanding of Large Language Models like BERT, GPT-3, and ChatGPT. Grazer Philosophische Studien, 99(4).
• Gubelmann, R. (2024). Pragmatic Norms Are All You Need – Why The Symbol Grounding Problem Does Not Apply to LLMs. In Y. Al-Onaizan, M. Bansal, & Y.-N. Chen (Eds.), Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (pp. 11663–11678). Association for Computational Linguistics.
• Harnad, S. (1990). The symbol grounding problem. Physica D: Nonlinear Phenomena, 42(1), 335–346.

The Symbol Grounding Problem (SGP) questions how symbols in AI systems can be meaningfully connected to real-world entities. The philosophical version of SGP emerges from the Computational Theory of Mind (CTM) which suggests that for AI to truly "understand," its symbols must be grounded in reality. This topic addresses the debate on whether the philosophical Symbol Grounding Problem (SGP) applies to large language models (LLMs). The focus is on whether LLMs inherently lack meaning due to the absence of explicit grounding or whether LLMs can meaningfully "understand" language through pragmatic norms alone, without relying on explicit symbol grounding in the real world.

Potential Research Questions:

• What does it mean for an LLM to "understand" linguistic meaning? Why is it even important?
• How might pragmatic theories of meaning offer a more accurate conceptual framework than correspondence theories for evaluating LLM language comprehension?
• Can AI systems operate purely on "pragmatic norms" without real-world reference?
• Could pragmatic norms alone allow an AI system to understand and translate languages without ever observing explicit real-world grounding references?

Literature

• Bender, E., & Koller, A. (2020). Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.
• Gubelmann, R. (2024). Pragmatic Norms Are All You Need – Why The Symbol Grounding Problem Does Not Apply to LLMs. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 11663–11678.
• Pavlick, E. (2023). Symbols and grounding in large language models. Philosophical Transactions of the Royal Society A, 381(2251), 20220041.
• Emily Bender and Alexander Koller. 2020. Climbing towards nlu: On meaning, form, and understanding in the age of data. Proceedings ofthe 58th Annual Meeting ofthe Association for Computational Linguistics, pages 5185–5198.
• Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28(1), 3–71.
• Glüer, K., Wikforss, Å., & Ganapini, M. (2024). The Normativity of Meaning and Content. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (Fall 2024). Metaphysics Research Lab, Stanford University.
• Gubelmann, R. (2023). A Loosely Wittgensteinian Conception of the Linguistic Understanding of Large Language Models like BERT, GPT-3, and ChatGPT. Grazer Philosophische Studien, 99(4).
• Gubelmann, R. (2024). Pragmatic Norms Are All You Need – Why The Symbol Grounding Problem Does Not Apply to LLMs. In Y. Al-Onaizan, M. Bansal, & Y.-N. Chen (Eds.), Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (pp. 11663–11678). Association for Computational Linguistics.
• Harnad, S. (1990). The symbol grounding problem. Physica D: Nonlinear Phenomena, 42(1), 335–346.

Digital transformation is one of the dominant phenomena of our time and is "here to stay" (Carroll et al., 2023a). Profound changes due to the increasing use of digital technologies in all areas of life are taking place across levels and are transforming organizations, industries, and society as a whole (Ismail et al., 2017). Despite the high level of attention in academia and practice, digital transformation as a complex and multifaceted phenomenon is still not fully understood, and heterogeneous conceptualizations and definitions exist (Kraus et al., 2022; Vial, 2019). 

Gaining a deeper understanding of the drivers, mechanisms, processes, and outcomes of digital transformation is crucial for ensuring long-term success in digital transformation processes and engages academia and practitioners alike. Since digital transformation is not a static phenomenon, but can be viewed as a dynamic process that evolves over time (Carroll et al., 2023b), it is crucial to gain a deeper understanding of the drivers and outcomes of digital transformation processes to ensure their long-term success. 

A typical focus is on understanding and designing the digital transformation of an organization (Vial, 2019). However, digital transformations increasingly take place at an interorganizational or industry level (Mann et al., 2022; Plekhanov et al., 2023). Understanding digital transformation on the industry level is crucial for policymakers and stakeholder organizations alike, as macro-environmental aspects like industry trends have a strong impact on organizations (Mergel et al., 2019).

As part of this seminar paper, this research topic will be explored empirically using a qualitative approach. To this end, semi-structured expert interviews will be conducted with interview partners who are dealing with digital transformation in their industries.

Literature

• Carroll, N., Conboy, K., Hassan, N. R., Hess, T., Junglas, I., & Morgan, L. (2023b). Problematizing assumptions on digital transformation research in the information systems field. Communications of the Association for Information Systems, 53(1), 508-531.
• Carroll, N., Hassan, N. R., Junglas, I., Hess, T., & Morgan, L. (2023a). Transform or be transformed: the importance of research on managing and sustaining digital transformations. European Journal of Information Systems, 32(3), 347-353.
• Ismail, M. H., Khater, M., & Zaki, M. (2017). Digital business transformation and strategy: What do we know so far. Cambridge Service Alliance, 10(1), 1-35.
• Kraus, S., Durst, S., Ferreira, J. J., Veiga, P., Kailer, N., & Weinmann, A. (2022). Digital transformation in business and management research: An overview of the current status quo. International journal of information management, 63, 102466.
• Mann, G., Karanasios, S., & Breidbach, C. F. (2022). Orchestrating the digital transformation of a business ecosystem. The Journal of Strategic Information Systems, 31(3), 101733. 
• Mergel, I., Edelmann, N., & Haug, N. (2019). Defining digital transformation: Results from expert interviews. Government information quarterly, 36(4), 101385.
• Plekhanov, D., Franke, H., & Netland, T. H. (2023). Digital transformation: A review and research agenda. European Management Journal, 41(6), 821-844. 
• Vial, G. (2019). Understanding digital transformation: A review and a research agenda. The Journal of Strategic Information Systems, 28(2), 118-144.

Having understood the significant effects of artificial intelligence (AI) on operational efficiency (Cui et al., 2024), organizations strategically focus on deploying and providing artificial intelligence in their processes and structures. Organizational members across departments now routinely and increasingly interact with conversational AI to reduce highly repetitive tasks and focus more on higher-level cognitive tasks at work, with 88% of AI users coming from non-technical professions (De Smet et al, 2023). AI interactions are likely to become even more common as generative AI is increasingly used in roles and functions previously reserved for humans, such as HR, IT, finance, or customer service and support (Tey et al, 2024). Estimations of the World Economic Forum indicate that until 2020, AI-driven automation will autonomously take over one-third of all work tasks (Di Battista et al., 2025). 

The use of GenAI brings implications for the working environment that are important today and in the future to generate hoped-for efficiencies: the way of working through the use of AI in business processes is changing and interpersonal collaboration in teams is being influenced as AI is used alongside human colleagues for monitoring, coordination and operational work. With AI being increasingly embedded in collaborative processes, this technology challenges, the understanding of the technology itself (Larson & DeChurch, 2020), traditional notions of teamwork (Richter & Schwabe, 2025), and intragroup processes (Zercher et al. 2023). 

The increasing application and use of AI have a significant impact on socio-technical work systems. In particular, challenges and requirements for leaders and leadership itself can be identified, highlighting the critical role of leadership for successful implementation (Mayer et al., 2025) and adoption of AI in the organization (Pfeifer et al, 2022). Current studies indicate that the presence of AI in the work environment affects what expectations are placed on leadership e.g., managers must be able to listen and persuade more than decide and instruct (Watson et al, 2021), and, what responsibilities and tasks leaders have, e.g., leaders need to lead organizational AI initiatives to achieve the desired benefits while avoiding the negative consequences (Berente et al. 2021).

As part of this seminar paper, a systematic literature review / qualitative investigation will be conducted to assess the state of the art on academic literature on the influence of GenAI on leadership tasks based on Mintzberg (1973).

Literature

• Berente, N., Gu, B., Recker, J., & Santhanam, R. (2021). Managing artificial intelligence. MIS quarterly, 45(3).
• Cui, et al. (2024). The effects of generative ai on high skilled work: Evidence from three field experiments with software developers. Available at SSRN 4945566.
• De Smet, A., Durth, S., Hancock, B., Baldocchi, M., & Reich, A. (2023). The human side of generative AI: Creating a path to productivity. McKinsey & Company
• Di Battista, A., Grayling, S., Játive, X., Leopold, T., Li, R., Sharma, S., & Zahidi, S. (2025). Future of jobs report 2025. In: World Economic Forum, Geneva, Switzerland. www.weforum.org/publications/the-future-of-jobs-report-2025/digest/.
• Larson, L., & DeChurch, L. A. (2020). Leading teams in the digital age: Four perspectives on technology and what they mean for leading teams. The leadership quarterly, 31(1), 101377.
• Mayer, H., Yee, L., Chui, M. & Roberts, R. (2025), Superagency in the workplace: Empowering people to unlock AI’s full potential. McKinsey & Company
• Mintzberg, H. (1973). The Nature of Managerial Work. New York, Harper & Row.
• Peifer, Y., Jeske, T., & Hille, S. (2022). Artificial intelligence and its impact on leaders and leadership. Procedia computer science, 200, 1024-1030.
• Richter, A., & Schwabe, G. (2025). “There is No ‘AI’in ‘TEAM’! Or is there?”–Towards meaningful human-AI collaboration. Australasian Journal of Information Systems, 29.
• Tey, K. S., Mazar, A., Tomaino, G., Duckworth, A. L., & Ungar, L. H. (2024). People judge others more harshly after talking to bots. PNAS nexus, 3(9), pgae397.
• Watson, G. J., Desouza, K. C., Ribiere, V. M., & Lindič, J. (2021). Will AI ever sit at the C-suite table? The future of senior leadership. Business Horizons, 64(4), 465-474.
• Zercher, D., Jussupow, E., & Heinzl, A. (2023). When AI joins the Team: A Literature Review on Intragroup Processes and their Effect on Team Performance in Team-AI Collaboration. ECIS 2023 Research Papers. 307. 

The Process Innovation and Automation (PIA) Lab is a newly established learning and research environment that connects students, academics, and industry partners in exploring the future of process innovation, and automation. Teaching cases are a powerful method to capture the dynamics of such labs, making them accessible for both educational and research purposes. By systematically documenting the perspectives of different stakeholders, a teaching case can provide students with a realistic and multi-layered understanding of opportunities and challenges in digital transformation.

The seminar paper aims to develop a teaching case about the PIA Lab. Students are expected to conduct interviews with three different stakeholder groups and integrate their perspectives into a structured teaching narrative. The final output should outline the Lab’s objectives, usage scenarios, perceived benefits, and challenges, framed in a way that enables teaching and classroom discussion.

In addition, the seminar paper should follow a Design Science Research (DSR) approach. The teaching case is to be treated as a designed artefact that is both constructed (based on stakeholder insights) and evaluated (regarding its usefulness for teaching and learning). Students are therefore expected not only to design the case narrative but also to reflect on its evaluation, e.g., by formulating learning objectives, designing guiding questions for class discussions, and considering potential improvements.

• Students – who experience the Lab as part of their learning journey.
• Lecturers – who use the Lab for teaching and didactic experimentation.
• Industry partners – who engage with the Lab for applied research, prototyping, or training.

Literature

• Tuunanen, T., Winter, R., vom Brocke, J. (2024). Dealing with Complexity in Design Science Research: A Methodology Using Design Echelons. MIS Quarterly,48(2), 427–458.
• Ellet, W. (2018). The case study handbook: A student’s guide. Harvard Business Press.
• Venable, J., Pries-Heje, J., Baskerville, R. (2016). FEDS: a Framework for Evaluation in Design Science Research. European Journal of Information Systems, 25(1), 77–89.
• Yin, Robert K. Case study research and applications. Vol. 6. Thousand Oaks, CA: Sage, 2018.