As machine learning (ML) instruction is dramatically widened its audience, new tools and methods are needed to allow an effective comprehension also by students without a strong mathematic/programming background, who may have more difficulty in districting among the fundamental theoretical concepts. To address this gap, we propose MLopoly, a Monopoly-inspired serious game designed to support interactive and experiential learning of core ML topics. The game maps decision-making challenges across realistic scenarios (e.g., restaurant, farm, bank) where players iteratively build, tune, and evaluate ML models. Game mechanics such as property upgrades, rent calculations, and challenge cards are directly tied to players’ performance on ML tasks. We discuss game design, gameplay dynamics and implementation issues, A preliminary validation study has been carried out conducting a 20-item questionnaire based on Bloom’s Taxonomy and Attention, Relevance, Confidence, and Satisfaction (ARCS) Motivation Model to assess learning and motivation. Results indicate that MLopoly offers an engaging, low-barrier environment for understanding and applying ML, also encouraging disciplinary deepening of the covered topics.