Bank of America ties 360-basis-point efficiency gain to AI

Bank of America reports measurable efficiency gains from enterprise AI deployment

Bank of America reported second-quarter results that CFO Alastair Borthwick directly attributed to scaled artificial intelligence deployment, marking one of the clearest attempts by a major financial institution to quantify AI's impact on core operating metrics.

The bank's efficiency ratio—operating expenses divided by revenue—improved 360 basis points year-over-year to 59%, down from 63% in the prior-year period. Revenue reached approximately $31.6 billion, net income hit $9.1 billion, and earnings per share came in at $1.21. Return on tangible common equity stood at roughly 17%.

Borthwick said more than 200,000 Bank of America employees now use AI capabilities that enable more effective work. He framed the efficiency improvement as evidence that AI investment is translating into measurable productivity rather than serving solely as a cost-reduction lever or benefiting from interest-rate tailwinds.

Why it matters

Most large enterprises remain stuck in AI pilot programs with limited visibility into financial returns. Bank of America's willingness to tie a specific efficiency metric to AI deployment—and to reinvest productivity gains in growth hiring rather than banking all savings—offers a concrete playbook for CFOs under pressure to justify AI spending. The approach also signals that competitive advantage may accrue to organizations that scale AI across tens of thousands of employees, not just specialized teams.

Consumer banking posts fifth straight quarter of positive operating leverage

The bank's consumer division generated $3.3 billion in quarterly earnings, up 10%, while achieving a 51% efficiency ratio and 29% return on capital. Revenue in the segment rose 5%. Holly O'Neill, president of consumer, retail and preferred banking, delivered a fifth consecutive quarter of positive operating leverage—revenue growth outpacing expense growth—while deploying new AI tools and investing in the branch network.

Borthwick pointed to Erica for Employees, an internal AI assistant, as a key productivity driver. The tool handles routine service requests and surfaces information for staff in branches and operations, allowing employees to focus on exceptions and higher-value interactions. He described the pattern as a gradual shift in how much human labor the bank requires per dollar of revenue, rather than one-time headcount reductions.

Middle-market investment banking expansion funded by efficiency gains

Bank of America announced multiple senior hires in regional investment banking on July 14, adding expertise in Austin, Boston, Charlotte, Chicago, Detroit, Minneapolis, New York, San Francisco, and West Palm Beach. The regional investment banking team has grown to more than 200 bankers across 26 U.S. cities since launching in 2016, reporting to co-heads Neil Kell and Samardh Kumar.

Borthwick characterized the hiring as a deliberate allocation of the efficiency dividend: deploying expensive front-office talent in pursuit of fee income growth while maintaining margin discipline through an AI-enabled core. The strategy bets that automation in back- and middle-office functions can subsidize relationship-driven revenue expansion in the U.S. middle market.

Three lenses for CFOs benchmarking AI programs

Borthwick's framing suggests finance leaders evaluating their own AI investments should track three dimensions: movement in the efficiency ratio as AI scales, a detailed scorecard covering costs, cycle times, and human-in-the-loop behavior, and explicit decisions about what share of productivity gains will fund growth investments versus flow to margin.

Bank of America's approach assumes AI-driven productivity can support simultaneous margin improvement and reinvestment. Whether that thesis holds across multiple quarters will determine if other large institutions adopt similar frameworks.

Details were first reported by Sheryl Estrada in Fortune's CFO Daily newsletter on July 15, 2026.