In a recent commentary that has quickly become a focal point for discussions about the intersection of artificial intelligence and financial stability, Torsten Slok, chief economist at Apollo Global Management, warned that the rise of sophisticated AI‑driven financial agents could dramatically accelerate the migration of consumer funds away from traditional, low‑interest deposit accounts. While the notion of a bank run conjures images of frantic crowds lining up at teller windows, Slok’s warning is rooted in a more subtle, technology‑enabled process: algorithms that continuously scan the market for the best possible yield and automatically reallocate cash in real time. If left unchecked, such behavior could create a feedback loop that squeezes the liquidity of banks that rely heavily on inexpensive checking and savings deposits to fund loans and other assets.
### How AI Agents Operate Modern AI agents, often built on large language models or advanced reinforcement‑learning frameworks, are increasingly being integrated into personal finance platforms, robo‑advisors, and even standalone budgeting apps. These agents are capable of ingesting a wide array of data—interest rates, inflation expectations, credit spreads, and macroeconomic indicators—then making recommendations, or even executing transactions, on behalf of users. In many cases, the agents are set to act autonomously, moving funds from a user’s checking account to a higher‑yield money‑market fund, a short‑term bond ETF, or a peer‑to‑peer lending platform the instant a more attractive rate appears. The key differentiator between traditional human‑driven financial decisions and these AI‑driven moves is speed and scale.
A human might check rates weekly or monthly, but an AI can evaluate them every minute, or even every second, across millions of accounts simultaneously. This capability means that when a new, higher‑yield product enters the market, the AI can instantly shift a substantial portion of the aggregate cash pool, draining the low‑interest deposits that banks have traditionally counted on for stable funding. ### Potential Consequences for Banks Banks have long relied on the so‑called “deposit‑funding advantage.” Checking accounts and basic savings accounts typically pay very low rates—often below the prevailing inflation rate—yet they provide banks with a cheap, stable source of capital. This capital is then used to issue mortgages, business loans, and other credit products that generate higher returns.
If AI agents begin to systematically pull funds from these accounts, banks could face a rapid contraction in their funding base. Unlike a classic bank run, which is driven by panic and often results in a sudden surge of withdrawals at a physical branch, an AI‑driven exodus would be a continuous, algorithmic outflow, potentially occurring overnight. The ramifications extend beyond the immediate liquidity strain. A reduction in low‑cost deposits forces banks to seek alternative funding sources, such as issuing higher‑interest certificates of deposit (CDs) or turning to the wholesale funding markets, where borrowing costs are typically higher and more volatile.
This shift can compress net interest margins, reduce profitability, and ultimately lead banks to tighten credit standards—an outcome that could ripple through the broader economy. ### Regulatory and Market Responses Regulators are already aware of the broader challenges posed by rapid technological change in finance, but the specific threat of AI‑driven deposit migration is relatively new. Potential policy responses could include: 1. **Enhanced Liquidity Requirements**: Requiring banks to hold a larger buffer of high‑quality liquid assets to absorb sudden outflows.
2. **Transparency Obligations**: Mandating that AI‑driven platforms disclose the extent to which they automatically move customer funds and the criteria they use.
3. **Rate Caps on Automated Transfers**: Setting limits on how frequently or how much cash can be moved without explicit user confirmation. 4. **Collaboration with FinTech**: Encouraging banks to partner with fintech firms to embed AI agents within their own platforms, thereby retaining control over the flow of deposits.
In addition to regulatory measures, banks themselves can adopt proactive strategies. By offering competitive, tiered interest rates that reward longer‑term balances, or by bundling value‑added services such as financial planning tools, they can make their deposit products more attractive to both human and algorithmic decision‑makers. Some institutions are already experimenting with “smart” accounts that integrate AI recommendations directly, allowing the bank to retain the relationship while still offering the best possible returns.
### Broader Economic Implications If AI agents become the dominant force in allocating household cash, the traditional banking model could undergo a fundamental transformation. The speed and efficiency of AI‑driven reallocation could lead to a more optimal distribution of capital across the economy, directing funds toward higher‑yielding, potentially more productive uses.
However, the transition carries risks. A sudden, large‑scale shift away from bank deposits could exacerbate credit tightening at a time when the economy may already be vulnerable, for example during a slowdown or an inflationary period. Moreover, the concentration of decision‑making within a few large AI platforms raises concerns about market concentration and systemic risk.
If a handful of algorithms dominate the flow of capital, any technical glitch, cyber‑attack, or erroneous parameter change could trigger a cascade of withdrawals, amplifying the very risk that Slok warns about. ### Conclusion Torsten Slok’s cautionary note serves as an early warning that the convergence of AI technology and everyday banking practices could reshape the landscape of deposit funding. While the promise of higher returns for consumers is appealing, the systemic implications of automated, large‑scale cash movements cannot be ignored.
Stakeholders—including banks, fintech developers, regulators, and consumers—must work together to design safeguards that preserve liquidity stability while still leveraging the efficiency gains offered by AI. By anticipating these dynamics now, the financial system can adapt responsibly, ensuring that the benefits of AI are realized without compromising the foundational stability of banks that have long been the bedrock of modern economies.