Over the past decade, artificial intelligence (AI) has been increasingly deployed in corporate operations. However, its impact on mitigating damages from natural disasters—a major source of uncertainty—has been largely overlooked in practice. We empirically investigate whether, and through what mechanisms, AI enables firms to enhance their resilience in the face of disaster shocks. In particular, we examine which types of firms are better equipped to harness the full potential of AI.

We compiled and analyzed a comprehensive dataset of disasters and identified firm-specific exposures at a highly granular county level. Utilizing stock return dynamics as a proxy for firm recovery, we find that AI indeed mitigates the adverse effects of disasters. In our exploration of underlying mechanisms, we discover that AI alleviates the declined labor responsiveness during volatile periods. Given that timely labor adjustments—such as employee reallocation or recruitment—are particularly challenging in the context of disasters compared to adjustments in financial capital, the operational mechanisms of AI present a promising remedy for enhancing the effectiveness of constrained factors.

Furthermore, our analysis indicates that firms with higher organizational complements—characterized by great IT accumulation, a more educated workforce, and increased use of AI in decision-making roles—are better positioned to leverage the mitigating effects of AI when confronted with uncertainty, rather than firms that excel in traditional performance metrics such as profitability or cash reserves.