The competition between human and AI streamers: live streaming strategies in a duopoly market considering consumer heterogeneity

AI streamers are increasingly prevalent in e-commerce live streaming, yet their competitive impacts and the factors driving enterprise psychological decisions remain underexplored. This paper explores the decision-making process of e-commerce enterprises in adopting streamers and the sensitive factors affecting their decisions from the perspective of heterogeneous consumers. Based on the SOR theory, this study constructs static and dynamic game models within a duopoly market, using the framework of "streamers' stimuli-heterogeneous consumers' utility perception-decision making process of streamers selection" and finds that: (1) The market composition of rational and impulsive consumers does not affect enterprises' pricing, supply or profits, while loyal consumers impact profit and decision speed without altering system equilibrium; (2) When the customer stickiness and fan base of the two oligarchs are the same, using AI streamers can increase profits, and enterprises with higher consumer engagement are more likely to adopt this strategy; (3) Live streaming efficacy and marketing costs are critical factors shaping supply, pricing and profitability, particularly in asymmetric models. Cost disparities can guide strategies and maintain system stability within specific thresholds; (4) Consumer trust and interaction quality not only impact marketing strategies and system stability, but also positively influence adopters' supply and suppress competitors' market share; (5) Enterprise preference for streamers influences competitors' strategies and convergence rates. Enterprises prefer AI streamers with strong learning abilities and high consumer satisfaction, as they maximize profits. This study offers theoretical guidance for e-commerce enterprises to adopt AI streamers and formulate effective market strategies.