Too little, too much, or just right? How AI recommendation information volume shapes trust and consumer behavior

Artificial intelligence (AI) shopping assistants explain why a product is selected, yet the appropriate defaults remain unclear. We define recommendation information volume as the amount of nonredundant, product-relevant information and compare five ordered explanation packages. Study 1 identifies an inverted-U response in a randomized Chinese mobile-retail experiment. Study S1 compares identical messages attributed to an AI assistant or a human salesperson, with separate source-neutral measurement validation. Study 2 reports conditional positive indirect associations through perceived usefulness, trust in AI, and decision comfort, and conditional negative indirect associations through information overload and algorithmic over-justification. Same-session estimates do not establish causal mediation. Its product-version comparison combines complexity, functionality, and safety considerations. Study 3 provides behavioral evidence across 51,248 recommendation impressions and service outcomes for 5000 distinct users. Moderate packages perform best within the tested sets. Bounded information processing explains the benefit–burden trade-off; persuasion knowledge suggests that disproportionate justification depends on expectations of selective versus comprehensive advice, across AI and human sources. Because amount, diagnostic priority, and content composition vary jointly, the experiments identify assigned-package effects, not pure quantity effects. The findings support category-specific testing of default explanations without establishing an exclusively algorithmic mechanism.