This paper aims to advance understanding of value co-creation and value co-destruction in artificial intelligence (AI)-enabled services by introducing the Refined Interaction Value Framework (R-IVF). The R-IVF explains how interaction value depends not only on technical performance, but also on how users interpret machine initiative, calibrate trust, experience meaningful exploration and evaluate contextual fit.
This paper develops the R-IVF conceptually by integrating three AI-era constructs – Agentic Perception (AP), Dynamic Trust Calibration (DTC) and Serendipitous Exploration (SE) – into the original framework and by reconceptualizing Configuration Value as Contextual Intelligence, an integrative mechanism that situates these constructs within the user’s evolving context. The framework is illustrated through two contrasting AI-enabled service contexts and extended through cross-functional managerial implications and an illustrative diagnostic monitoring approach.
The paper shows that interaction value in AI-enabled services is shaped by three recurrent tensions: increasing machine agency within interaction, the need for dynamically calibrated trust and the challenge of preserving meaningful exploration under increasingly personalized service conditions. The R-IVF explains how these tensions may either support value co-creation or contribute to value co-destruction, depending on whether AI-enabled interactions are experienced as supportive, trustworthy, contextually appropriate and open to meaningful discovery.
The R-IVF offers managers a practical lens for designing, governing and monitoring AI-enabled service interactions. It highlights cross-functional priorities across design, operations and customer communication and supports the use of illustrative diagnostic indicators to identify emerging vulnerabilities such as overreach, trust erosion, constrained exploration and contextual mismatch.
This paper extends interaction value theory into the AI era by introducing AP, DTC and SE as AI-specific mechanisms shaping service interaction and by reconceptualizing Configuration Value as Contextual Intelligence. In doing so, it reframes interaction value as an adaptive, context-sensitive process rather than a fixed set of service attributes. It contributes a conceptually grounded framework that clarifies how AI-enabled interactions may sustain, undermine, or reshape value over time.
