Live-streaming E-commerce operations: is it wise for manufacturers to use AI streamers when human streamers face a goodwill crisis?

In addition to selling products directly to consumers in traditional online channels through an e-platform at a traditional online price, a manufacturer sells products in live-streaming channels by employing a human streamer. Human streamers with goodwill can attract additional consumers (fan size). However, they may make erroneous statements during live-streaming sales, triggering a goodwill crisis that damages product sales. In response, some manufacturers invest in artificial intelligence (AI) streamers to support product sales in traditional online channels, but at investment costs. However, most existing literature focuses on the human streamer’s fan size and rarely considers goodwill crises. Therefore, we develop a game-theoretic model to examine the impact of goodwill crises and further explore manufacturers’ preferences for human and AI streamers. The main findings are as follows. Intuitively, when the crisis damage degree is high, manufacturers invest in AI streamers. However, we find that only when the human streamer’s fan size is low, manufacturers choose AI streamers that are profitable if the crisis damage degree is high. Otherwise, manufacturers still choose human streamers. Further, manufacturers investing in AI streamers can increase the e-platform’s profit.