Unveiling the role of Artificial Intelligence in Search Strategy: a Patent analysis through Natural Language Processing

This study investigates the impact of Artificial Intelligence (AI) adoption on firms' exploration and exploitation strategies in innovation. Amidst a rapidly evolving competitive landscape, firms must innovate to survive, either through incremental refinement of existing capabilities (incremental innovation) or by making significant leaps (radical innovation). The ability of AI to process diverse and voluminous data sets suggests it could significantly shift the traditional balance between exploration and exploitation, potentially redefining strategic outcomes in terms of innovation efficiency and firm performance. To investigates the impact of AI on firms' exploration and exploitation strategies, we need to identify firms that have adopted AI technologies and assess their levels of exploration and exploitation. We use patent activity as a proxy to determine AI adoption. Whereas, to understand the level of exploration and exploitation of firms we analysed the text of patents using Natural Language Processing. We categorize firms into those with AI-related patents and those without, employing a difference-in-difference methodology to explore changes in their innovation search strategies over time. This approach helps to isolate the impacts of AI on firms' strategic behaviors in terms of exploration and exploitation. Preliminary findings indicate that AI adoption correlates with a marked increase in explorative activities, suggesting that AI not only supports but actively promotes exploration over exploitation.