Why do some manufacturing firms sustain product innovation under conditions of environmental uncertainty, digital disruption and resource constraints? We approach this question by arguing that biomimicry-inspired capabilities (BIM) enhance product innovation capability (PIC) by enabling artificial intelligence capability (AIP) to translate ecological insights into scalable innovation outcomes, while adaptive resource capacity (ARC) conditions the effectiveness of this process. Drawing on the resource-based view and dynamic capabilities theory, we conceptualize BIM as a problem-framing capability, AIP as a process-enabling mechanism and ARC as a resource orchestration capability that shapes innovation execution.
We test our arguments using a sequential mixed-method design. First, we analyze survey data from 318 manufacturing firms using PLS-SEM to examine the proposed moderated-mediation relationships. Second, we draw on 21 semi-structured interviews to uncover the underlying process mechanisms and contextualize the quantitative findings.
Our findings indicate that BIM positively influences both AIP and PIC, with AIP mediating the translation of ecological inspiration into innovation outcomes. Moreover, ARC exhibits a paradoxical moderating effect: while moderate resource flexibility strengthens the BIM–AIP–PIC pathway, excessive slack weakens process discipline and reduces innovation effectiveness.
In this way, we reveal that the value of artificial intelligence and resource flexibility is contingent on how they are structured and deployed within innovation processes. The study contributes by showing that innovation is not driven by isolated capabilities but by their coordinated orchestration and by identifying the limits of resource flexibility in sustaining innovation under dynamic conditions.
