The AI Productivity Dividend Just Hit the P&L (And Most Banks Aren’t Ready)
Bank of America eliminated 1,000 jobs in Q1 2026. Same quarter: 290 basis points of operating leverage, 10% increase in technology spending, and active hiring for AI engineers and agentic AI leads.
CEO Brian Moynihan didn’t hedge. The cuts were the result of “eliminating work and applying technology.”
This is what AI productivity looks like when it moves from the innovation lab to the income statement. Not a pilot. Not a proof of concept. Measurable operating leverage, flowing to shareholders, funded by automation that is already working.
And it’s not isolated. Anthropic’s Mythos model launched April 7 under Project Glasswing, a restricted-access program for frontier agentic AI. Within two weeks, regulators in Australia, Singapore, and Hong Kong issued formal supervisory risk statements. Yet Piraeus Bank and Accenture launched a production AI Hub using Anthropic’s Claude architecture anyway, explicitly framing it as a necessity in “the global AI race.”
The cost of waiting for regulatory clarity now exceeds the cost of moving under uncertainty.
Meanwhile, Citi posted three senior AI engineering roles in one week, all emphasizing autonomous systems and production deployment. FIS and Barclays extended their core partnership, with Barclays US migrating to the cloud-ready FIS Profile platform. Visa is positioning Pismo as the incremental modernization path that avoids rip-and-replace risk.
The banks treating workforce transformation, AI governance, and core modernization as strategic imperatives (not IT projects) are separating from the pack right now.
Subscribe to BIS, the Banking Intelligence Service from Core System Partners, for the full breakdown including CB Radar vendor tracking, Rick’s Strategic Take on talent architecture as competitive advantage, regulatory tempo analysis, and the Mythos deployment framework, delivered weekly. Banking Intelligence Service
