Goldman Sachs has determined that the advancement of artificial intelligence broadly benefits the credit fundamentals of the banking sector, with its impact operating through two primary channels. First, the AI investment cycle generates increased financing and fee income for banks. Second, banks themselves are deploying AI to elevate automation levels, reducing costs and enhancing operational efficiency.

To begin with, AI infrastructure construction demands substantial capital investment, creating new business opportunities for banks. As data centers, computing infrastructure, and related enterprises expand their spending, banks can engage in providing merger and acquisition advisory services, bond and equity underwriting, loan syndication arrangements, asset management, and direct lending. Notably, large-scale AI-related projects typically require structured financing solutions.

Banks can initially offer construction loans, bridge financing, and project finance before distributing longer-duration risk to institutional investors or private market funds, thereby capturing loan spreads and fee income. Goldman Sachs points out that the frequency with which US banks discuss AI capital expenditure on earnings calls has become markedly higher than that of non-financial corporations, reflecting that banks view AI infrastructure expansion as a fresh commercial opportunity.

Lending attitudes are also undergoing a transformation. The Federal Reserve's January 2026 Senior Loan Officer Opinion Survey indicates that banks are exhibiting a greater willingness to lend to businesses anticipated to benefit from AI development. Goldman Sachs projects that if capacity constraints tighten in the traditional syndicated loan market, banks may increase their participation in AI-related financing, leveraging their own balance sheets to provide funding directly to enterprises.

Beyond external business prospects, AI has the potential to directly improve banks' internal cost structures. Banks can utilize AI to handle back-office operations, data analysis, customer service, compliance, and process automation, thereby boosting employee productivity and strengthening cost control. Consequently, AI's influence on banks extends beyond mere new loan demand.

For large banking institutions, income can be derived from both sides of the equation. On one hand, they can capture investment banking, financing, and asset management revenues from the AI capital expenditure cycle. On the other hand, they can reduce operational costs through internal AI deployment, with both factors jointly enhancing profitability. This shift is also becoming apparent in banks' own funding activities.

Goldman Sachs observes that recent US dollar bond issuance by American banks has increased, while euro-denominated supply from European investment-grade banks has remained relatively stable overall. The uptick in US bank dollar bond issuance is linked to banks proactively locking in financing costs when credit spreads are narrow, and it also mirrors increased balance sheet utilization, including funding for prime brokerage activities, commercial and industrial loans, and financing tied to the AI supply chain.

Goldman Sachs believes that as AI capital expenditure continues to expand, banks' roles across the financing spectrum could strengthen further. Banks can act as both loan capital providers and undertake underwriting, syndication distribution, asset management, and risk transfer functions, generating revenue from multiple stages of the AI investment cycle.

Overall, AI's impact on the banking industry is not simply a "technology-driven cost reduction" narrative. Goldman Sachs emphasizes that, more critically, AI simultaneously broadens banks' revenue streams and efficiency improvement potential. Externally, the AI investment cycle is creating new lending and fee-based businesses. Internally, AI promises to raise automation levels and compress operating costs. The combination of these two pathways positions AI as a progressively significant variable influencing bank profitability and credit performance.