BofA Adds 4,000 Staff to Fix AI-Found Security Flaws
Bank of America is adding 4,000 staff members to address security flaws found by artificial intelligence models. The company has deployed 1,000 contractors and 3,000 employees to patch software vulnerabilities uncovered by Anthropic’s Mythos and other frontier AI systems. The effort highlights the rising cost of AI-powered cybersecurity, as these models can expose weaknesses that banks previously did not know existed.
This creates a massive remediation workload for financial institutions trying to keep up with the speed at which AI can find new risks. The deployment of 4,000 people to address these flaws illustrates the scale of the problem, as the AI models can identify vulnerabilities that have existed for years or even decades, forcing banks to remediate decades of accumulated software weaknesses.
Banking on AI Safety
PNC Financial has signaled a shift in how deposits will be protected. The bank predicts deposit mobility will become increasingly frictionless, meaning customers can easily move their money to competitors. To maintain relationships, banks must match third-party technology without losing the service connection to the checking account. CEO William Demchak drew a firm line on autonomous AI transactions, describing agents acting inside customer accounts as “incredibly unsafe” without stronger liability rules, authentication, and human oversight.
The bank is investing in a new mobile app, rewards, and 55 new branches this year to make the core relationship strong enough that customers do not need outside agents. This investment strategy is a direct response to the threat of third-party agents managing customer funds, as PNC accepts that the old friction protecting deposits is disappearing.
U.S. Bancorp reported that credit card balances rose 11.5% year over year to $40.2 billion in the third quarter of 2026, while net charge-offs fell to 3.29%. Despite higher interest rates and softer consumer sentiment, spending remains resilient. Consumer deposits hit another quarterly record. However, the bank has not yet seen third-party AI agents automatically move customer cash between banks.
U.S. Bank believes the operational complexity currently keeps this threat at bay, giving institutions time to build their own optimization tools before the deposit relationship gets mediated by an agent. The bank is already thinking about what happens when agents start moving deposits, recognizing that if agents can compare rates and shift cash automatically, deposit stickiness becomes much harder to take for granted.
Intuit Expands into B2B Revenue Management
Intuit plans to acquire Sequence, an AI-native revenue management platform, to add pricing, billing, and collections capabilities to its Enterprise Suite. This move pushes Intuit deeper into the business-to-business sector, where pricing models are complex and usage-based. Revenue management is a difficult area for spreadsheets and generic accounting software.
Sequence gives Intuit another piece of the workflow, potentially giving growing businesses a reason to stay inside its ecosystem as their finances become more complicated. The acquisition expands Intuit deeper into B2B software and professional services, where usage-based, seat-based, and milestone-based pricing make revenue management a lot more complicated, and Intuit is moving further up the financial stack from helping businesses record what happened to helping them manage how they get paid.
BNY Mellon launched digital transfer agency capabilities that combine tokenization, distribution, and custody. This extends its fund-servicing business to digitally native funds across blockchains and jurisdictions. The bank also introduced Pay-to-Wallet, allowing financial institutions to send cross-border payments directly from bank accounts to participating retail digital wallets without building the infrastructure themselves.
BNY is taking a pragmatic route into digital assets by extending existing capabilities to new rails rather than building a separate crypto business. The strategy focuses on the infrastructure layer, making BNY more useful as financial activity moves onto new rails. The move also shows what BNY’s platform strategy looks like in practice: connecting existing businesses to solve new client problems rather than treating each new technology as a standalone business.
