AI is being increasingly used in corporate finance to equip businesses with more advanced forecasting capabilities, real-life scenario simulations, and investment risk quantification. However, the impact of AI decision support systems on managerial judgment has not yet been extensively researched. Based on the Ability – Motivation – Opportunity (AMO) model and the concept of behavioral finance, the paper constructs a conceptual model to test the interactions between the quality of the AI-assisted capital budgeting tool and managerial overconfidence and their relationship to investment efficiency. The model suggests that the use of high-quality capital budgeting tools improves managers' analytical skills, resulting in better investment decisions, but at the same time may result in the extent of their managerial overconfidence due to automation complacency, causing biased capital allocation. The relationship of overconfidence and investment efficiency is thought to be moderated by organizational slack; additionally, resource-rich firms have more room for overconfident investment. The paper proposes eight theoretical propositions on the topic, outlines the entire conceptual framework, and outlines specific mechanisms by which the use of AI tools helps or hinders cognitive bias in capital allocation decisions in business organizations. Theoretical work includes putting AI decision-support research into a behavioral corporate finance framework in the AMO framework, and addressing the conflicting effects of AI on judgment quality. The practical considerations provide recommendations for CFOs, boards of directors, and regulators looking to leverage the efficiency of AI and prevent it from fostering hazardous overconfidence. Empirical directions for future research provide the research methods for testing the propositions with survey, archival, and experimental research.