AI adoption in the presence of consumer resistance: is “win-win” possible for an AI supplier and a manufacturer?

Artificial intelligence (AI) is increasingly important in manufacturing supply chain operations. Considering its effects on consumer personalisation (CPE) and cost reduction (CRE), the manufacturer (M) is motivated to acquire licences from the AI supplier (S). However, high AI licence fees and consumer resistance due to data privacy concerns render M not reaping benefits from AI. Moreover, S may not benefit from AI because of the high development costs. To explore optimal strategic interactions between S and M and achieve “win-win” situations, we establish a game-theoretical model to characterise the equilibrium and analyse the impacts of the CPE, CRE, and consumer resistance on such “win-win” situations. Our findings show that when consumer product quality perception is low, “win-win” situations only arise if S’s AI development efficiency is high. Conversely, “win-win” situations occur only if both S’s development efficiency and M’s unit reducible cost are large. Finally, both AI’s CPE and CRE enhance M’s AI adoption and S’s motivation to raise its licence fee, promoting “win-win” situations. However, consumer resistance does not necessarily impair consumer utility, especially when consumers strongly prefer safeguarding their private data, but invariably impedes the “win-win” for M and S. This study offers strategic guidance for AI suppliers, manufacturers, consumers and policymakers navigating AI adoption in the presence of consumer resistance. For AI suppliers, enhancing AI’s cost reduction effect (CRE) and consumer personalisation effect (CPE) is crucial, as this enhancement justifies higher licence fees and facilitates “win-win” outcomes. However, AI suppliers must also address privacy concerns through transparent data practices to mitigate consumer resistance. For manufacturers, whether AI adoption is beneficial depends on not only coefficient of development cost of AI but also consumer quality perception and the unit reducible cost. Manufacturers should prioritise AI’s CPE when consumer quality perception is low and CRE when perception is high. For moderate consumer quality perception, the choice depends on the unit reducible cost. For consumers, the decision to embrace AI depends on how much they value their data privacy. If consumers care less about data privacy, embracing AI is the optimal strategy. Policymakers should encourage AI diffusion while safeguarding consumer data privacy, as such responsible adoption can improve social welfare even when consumer gains are limited. Alignment between supply chain strategies and regulatory frameworks is essential to optimise AI’s societal and economic benefits.