Not yet human-reviewed
This source record was written automatically by the research pipeline and has not yet been reviewed by the maintainer or integrated into a topic page. Treat it as a provisional summary and verify claims against the original source.
ChatGPT and Beyond: Exploring the Responsible Use of Generative AI in the Workplace
Citation. Söllner, M., Arnold, T., Benlian, A., Bretschneider, U., Knight, C., Ohly, S., Rudkowski, L., Schreiber, G., & Wendt, D. (2025). ChatGPT and beyond: Exploring the responsible use of generative AI in the workplace. Business & Information Systems Engineering, 67(2), 289–303. https://doi.org/10.1007/s12599-025-00932-8
Summary. This interdisciplinary discussion article frames responsible workplace GenAI use as a sociotechnical design problem spanning information systems, computer science, business psychology, law, and technology ethics. It argues that deployment must balance productivity and innovation against unreliable output, bias, copyright, privacy, autonomy, and well-being risks. The paper advances human oversight, role clarity, organizational policy and training, and GenAI literacy as conditions for preserving accountable human judgment.
Key claims
- Responsible workplace GenAI use requires workers to evaluate outputs for accuracy, bias, and copyright concerns, while managers establish acceptable-use policies, review processes, reporting protocols, and practical training.
- GenAI can reallocate work toward reviewing, revising, problem solving, and information processing; this may enrich jobs but can also raise mental demands and remove routine opportunities for recovery.
- A human-centered sociotechnical approach should evaluate productivity and decision quality alongside employee well-being, dignity, autonomy, privacy, and the ability to interpret, challenge, and override AI outputs.
- Delegating repetitive tasks can be useful, but the article argues that decision responsibility remains human because AI cannot bear normative responsibility and people may over-rely on apparently capable recommendations.
Evidence & limitations
- Peer-reviewed interdisciplinary discussion article drawing on a project group, workshops, conference discussions, and prior literature rather than reporting a new empirical study.
- The paper is open access under CC BY 4.0; its recommendations and research agenda should not be read as causal evidence that particular governance practices improve outcomes.