U.S. banks are investing unprecedented amounts in artificial intelligence, with annual spending at institutions like JPMorgan Chase, Bank of America, and Citigroup now exceeding $2–$4 billion each. Yet a critical national financial resilience problem persists: most U.S. banks cannot confidently determine whether these massive AI investments generate positive risk-adjusted returns, creating capital allocation inefficiency and potential systemic vulnerability. This research proposal outlines a comprehensive mixed-methods research design for investigating how senior executives in U.S. global banks govern enterprise AI investments, manage emerging financial risks, and measure return on investment when scaling AI across national banking operations. Drawing on the Resource-Based View, Paradox Theory, and the Technology-Organization-Environment framework, this proposal develops an integrated conceptual framework linking AI governance mechanisms, operating model configurations, and multi-dimensional ROI measurement specifically calibrated to the U.S. regulatory environment (Federal Reserve, OCC, FDIC). The proposed study would employ an embedded multiple-case design with semi-structured interviews of 30–40 C-Suite executives across 6–8 U.S.-headquartered global banks, supplemented by secondary analysis of SEC filings, FRED economic data, FDIC call reports, and Model Risk Management documentation. We propose a novel risk-adjusted ROI calculation framework incorporating direct financial benefits, indirect value creation, strategic option pricing, and probabilistic risk adjustments aligned with U.S. banking stress testing practices. Anticipated methodological barriers include organizational resistance, access constraints to senior executives, and causal attribution challenges—each addressed with specific mitigation strategies outlined in this proposal. This proposal aims to contribute empirically validated ROI measurement tools for executive decision-making at U.S. systemically important financial institutions and demonstrates a scholar-practitioner approach to bridging academic rigor with national financial stability priorities.