A review on algorithm aversion, appreciation, and investor return beliefs

As artificial intelligence (AI) continues to transform financial decision-making, responses of investors toward algorithmic tools have varied from rejection to voluntary adoption. This review looks at two different behavioral outcomes: algorithm aversion, or resistance to machine-provided advice even when it has been validated, and algorithm appreciation, where investors prefer algorithmic advice under certain particular conditions. Drawing on behavioral finance, psychology, and decision theory research, the review examines how such beliefs influence investor return beliefs i.e., the subjective investment performance expectations that people have. The review also examines the cognitive and affective processes underlying such beliefs, as well as the roles of trust, control, and framing in shaping investor attitudes. Findings show that institutional investors are more likely to algorithm appreciation with experience and data processing capacity, whereas retail investors have greater aversion with emotional bias and low transparency. The discussion is wrapped up with practical recommendations for improving algorithm acceptance, including higher user control, transparency in design, and blended advisory models. Closing the technical performance-user experience gap is paramount to encouraging effective, robust AI-driven investment systems.