AI Upskilling promoting Peer-to-Peer Learning using Self-Selected Real-Work Use Cases

Generative AI is transforming knowledge-intensive work and increasing the need for employees to develop skills for the responsible and effective use of generative AI. Companies therefore face challenges in supporting AI upskilling, including heterogeneous prior skill levels, difficulties in identifying use cases for AI‑supported work, and the need for continuous, context‑specific learning. This paper presents an AI upskilling approach based on work and learning projects (ALP), enabling employees to engage with generative AI by means of self‑selected real‑work use cases. The approach is based on work-integrated learning and combines structured methodological guidance with authentic work tasks to promote situated and transferable upskilling. The AI upskilling was implemented with employees at the Fraunhofer Institute for Industrial Engineering IAO, who selected real tasks from their daily work and processed them using a standardized sequence of learning steps. A dedicated website supported the process by providing tutorials, use case inspirations, and a prompt library for documenting and sharing prompts. Peer‑to‑peer learning sessions enabled employees to exchange experiences, validate emerging learnings, and collaboratively reflect on challenges encountered during AI‑supported work. The results indicate that combining embedded practice, self‑selected task processing, methodological scaffolding, and peer collaboration fosters the development of essential AI‑related skills, including prompting, critical curation of AI outputs, and contextualization. The AI upskilling approach supports scalable and sustainable skill development and demonstrates how work‑integrated learning formats can prepare employees to use generative AI responsibly and productively in dynamic, digitalized work environments.