Adapting teams to Human–AI collaboration: purposive programs and distinction-based task performance

Purpose This paper develops a systems-theoretical framework for understanding how organizations can strengthen team dynamics and strategic orientation in environments increasingly shaped by Generative Artificial Intelligence. It aims to clarify how purposive programming supports effective human–AI collaboration by drawing on empirical insights from an applied educational context. Design/methodology/approach The study adopts a conceptual approach grounded in Luhmann's organizational theory and is informed by practice-based workshop experiments in which student teams completed authentic tasks with and without GenAI support. The paper integrates purposive programming with distinction-based methods, systemic structural constellation formats, and the multifunctional organization concept to develop a structured orientation model for GenAI-enabled teamwork. Findings Results show that team processes – particularly shared goals, clear roles, mutual accountability, and communication quality – exert a stronger influence on performance than GenAI access itself. High-functioning teams using GenAI produced more creative and comprehensive outputs, whereas GenAI did not compensate for weak collaboration. GenAI primarily shifts the bottleneck from production to selection and validation. Research limitations/implications The educational and practice-based setting limits generalizability of the findings. Further research should assess the transferability of purposive programming across sectors. Practical implications A distinction-based four-stage pathway is proposed to support organizations in structuring GenAI-assisted teamwork through clear ends, explicit decision premises, and context-sensitive validation routines. Originality/value The paper links systems theory with GenAI-enabled collaboration and provides a strategic model for managing technological complexity.