AI-enabling investments and firm value: a contingent effect of systematic risk exposure

This is the authors' abstract. We don't add key points for this paper.

This study examines how systematic risk exposure shapes the valuation effects of AI-enabling digital investment, using panel data on Chinese SME and ChiNext listed firms from 2014 to 2023. We find that the marginal association between AI investment and firm value follows an inverted-U trajectory with respect to systematic risk. Value enhancement concentrates at moderate levels of non-diversifiable uncertainty, while the association turns negative at either end of the risk spectrum. The pattern is specific to systematic risk. Alternative proxies for idiosyncratic or earnings uncertainty do not reproduce it, consistent with real-options theory that ties the option-to-wait premium to non-diversifiable uncertainty. Evidence from the staggered roll-out of China’s AI Innovation Pilot Zones suggests that policy measures reducing investment irreversibility flatten this relationship. Organizational learning also attenuates the sensitivity. Inexperienced firms face substantial timing penalties that fade as they accumulate AI-investment experience.