Limits of AI-Enabled Value Creation - An explorative study on Artificial Intelligence in early-stage startups: implications for value creation, value capture, and dynamic capabilities

Abstract

This thesis investigates how artificial intelligence (AI) transforms business operations in early-stage startups. Although AI is widely accessible through commercially available tools, its broader implications for value creation, value capture, and organizational capabilities remain underexplored. To address this gap, the study adopts a qualitative research design based on semi-structured interviews with entrepreneurs, venture capitalists, and AI specialists in Silicon Valley and analyzes the data through an abductive approach. The findings show that AI is applied across a wide range of business activities and can be viewed through three mechanisms: replacing routine tasks, reinforcing existing processes, and revealing new opportunities. AI increases efficiency, lowers technical barriers, and accelerates product development. However, its role in enabling fundamentally novel business models is limited, as it primarily supports incremental innovation through the recombination of existing knowledge. The study also identifies key constraints and trade-offs in AI adoption, including the risk of hallucinations, dependence on external vendors, and issues related to data confidentiality and governance. As AI becomes embedded in workflows, it shifts roles toward orchestration, validation, and decision processes. This thesis contributes to theory in two ways. First, it shows that AI enhances value creation but constrains value capture due to vendor dependence and commoditization of capabilities. Second, it introduces three AI-Mediated Capabilities–AI Orchestration, Data Sovereignty, and Cognitive Control. Overall, the findings show that value creation and capture with AI depend not only on adoption but also on how firms govern the integration of AI into their business models.