Apollo Global Management's chief economist, Torsten Slok, warns that AI assistants could trigger an "agentic bank run" by automatically sweeping household cash out of low-yield checking accounts and into fintech products paying far higher rates. The risk is a rapid, coordinated drain of the cheap deposits banks rely on to fund lending, not a live event.
Apollo Global Management chief economist Torsten Slok has flagged a new kind of systemic risk at the intersection of artificial intelligence and consumer banking: what he calls the potential for an "agentic bank run." Slok says Muse and similar agentic AI assistants could soon automatically move household bank balances into higher-paying accounts, replacing the 0.1% that the average American household currently earns on checking deposits.
A wide gap between banks and fintech rates
Slok's note, published through Apollo's Daily Spark, asks whether an agentic bank run is coming if these tools go mainstream, sweeping cash into platforms paying 3.3% to 5.0% instead of the 0.1% national average on checking accounts. The FDIC national average sits at 0.4% for savings accounts and 0.1% for checking, according to the chart accompanying his note.
Eleven fintech products outpace traditional banks
An accompanying chart lists eleven fintech and online deposit products paying more than traditional banks, led by Adelfi at 5.0% and SoFi at 4.5%. LendingClub LevelUp sits at 4.2%, with Pibank at 4.1% and Revolut Metal and Current both at 4.0%. Fitness Bank and Varo sit at 3.8%, Wealthfront at 3.6%, Revolut Standard at 3.5% and AlumniFi at 3.3%.
Why banks would feel it as a systemic problem
Slok's underlying concern is not that one household chasing a better rate is a problem — it is what happens if that behavior becomes automatic and widespread at once. According to The Daily Hodl: "a problem for the entire financial system" is how Slok frames the risk if households broadly adopt AI agents to optimize cash returns. That kind of deposit base has traditionally been considered sticky, largely because most people do not actively manage small interest rate differences on transaction accounts — but an AI agent acting on a household's behalf removes that inertia.
The framing as a bank run, rather than simply a shift in deposit market share, reflects the systemic angle Slok is drawing: a fast, machine-driven reallocation of deposits away from banks could strain the funding models they use to support lending, with effects that extend beyond any single institution. Blockworks co-founder Jason Yanowitz highlighted the warning on X, calling the scenario fascinating.
Sources: Investinglive, Apollo Global Management
Trading involves risk.
Risk Disclaimer
