Catchword articles take up emerging topics of high relevance to both research and practice. They situate new concepts scientifically, structure the current state of knowledge, and identify future research needs.

The published article conceptualizes the still-nascent research field of agentic AI and distinguishes it from classical autonomous agents as well as multi-agent systems. The authors differentiate between individual AI agents and cooperating agent systems, define six central capabilities of modern agentic AI systems, and use a corporate example to show how multiple specialized AI agents can coordinate complex tasks across systems and—after human approval—autonomously execute business processes.

Beyond this, the researchers develop a research agenda addressing the topics of design, governance, human-AI collaboration, and platform ecosystems. The Catchword thereby provides a conceptual foundation for further research and the responsible deployment of agentic AI in enterprises.

Business & Information Systems Engineering (BISE) is among the internationally leading journals in information systems and publishes pioneering contributions at the intersection of information systems, digitalization, and management.

The full article is available at the following link: https://doi.org/10.1007/s12599-026-01009-w

Li, M.M., Reinhard, P. & Leimeister, J.M. Agentic AI. Bus Inf Syst Eng (2026). https://doi.org/10.1007/s12599-026-01009-w