Does AI Innovation Improve Corporate Investment Efficiency

This working paper examines whether artificial intelligence (AI) innovation improves corporate investment efficiency or may instead encourage excessive investment. It distinguishes between AI invention, adoption, corporate disclosure, and broader digitalization, recognizing that these activities capture different dimensions of corporate AI capability. The paper combines a targeted literature review with a proposed empirical blueprint for U.S.-listed nonfinancial firms. The intended baseline covers the period 2010–2024 and focuses on whether AI-related capabilities are associated with lower investment inefficiency, reduced underinvestment, or increased overinvestment. The proposed framework uses firm fundamentals to estimate expected investment and analyzes deviations from that benchmark, including CAPEX, R&D, acquisitions, and broader investment measures. Existing studies generally report favorable associations between AI-related measures and corporate investment efficiency. However, the evidence does not establish that AI patents cause better investment decisions across U.S. firms. The paper therefore emphasizes the distinction between association and causality, as well as the limitations of patent-based, disclosure-based, and policy-based measures of AI exposure. In addition to its academic contribution, the paper develops a practical capital-allocation framework for CFOs. It recommends selective investment, staged funding for uncertain applications, disciplined measurement of economic benefits, independent monitoring, and clearly defined expansion, pause, and termination criteria. No original firm-level panel has been assembled or estimated in this version. Accordingly, the paper reports no new regression coefficients, causal effects, confidence intervals, or project-level returns. All numerical illustrations are hypothetical. The contribution is a literature-based evidence synthesis, a reproducible empirical blueprint, and a managerial framework for evaluating AI-related corporate investment.