Financial regulators warn AI could raise risks in financial markets

In short: Financial policymakers are warning that wider use of AI could make financial markets less stable and, in some cases, less trustworthy.

What's going on

Andrew Bailey, the Bank of England governor and chair of the Financial Stability Board, warned G20 finance ministers about two main risks from advanced AI systems.

First, he said excitement about AI could inflate company values and increase “leverage” (borrowing to invest, like buying a house with a small down payment). If prices later fall, heavy borrowing can make the losses spread faster.

Second, Bailey warned that AI could make cyber attacks more effective and cheaper to carry out. He pointed to a particular worry: many firms depend on a small number of outside tech providers. That concentration means one serious failure could affect lots of companies at once, like a power outage that shuts down an entire neighborhood.

The Financial Times column also highlights a separate idea discussed at the Jackson Hole meeting of central bankers. Economist Markus Brunnermeier argued that “agentic AI” (AI that can take actions on its own, like a software worker following steps without being asked each time) may create “asymmetric understanding.” In simple terms, people might not be able to understand why AI systems make certain choices, even if those systems can predict how humans will react.

What to watch

If policymakers believe markets are becoming harder to trust or interpret, they may rely less on nudging expectations and more on direct rules, such as requiring banks to hold more reserves. Regulators may also push for more diversity in critical tech suppliers, so a single outage or attack does not freeze large parts of the system.

Source: Financial Times