WASHINGTON (Reuters) - Rapid adoption of artificial intelligence (AI) could create new risks for the U.S. financial system if the technology is not properly supervised, a panel of regulators warned on Thursday.
The Financial Stability Oversight Council, which comprises top financial regulators and is chaired by Treasury Secretary Janet Yellen, flagged the risks posed by AI for the first time in its annual financial stability report.
While the group said AI could spur innovation or efficiencies at financial firms like banks, the rapidly advancing technology requires vigilance from both the companies and their watchdogs.
"AI can introduce certain risks, including safety and soundness risks like cyber and model risks," the group said in its annual report published Thursday, adding it recommended firms and their regulators "deepen expertise and capacity to monitor AI innovation and usage and identify emerging risks."
The panel also flagged the growing role of nonbanks and private credit as meriting close attention, and said financial institutions and regulators should continue to try to better gauge risks stemming from climate change.
Some AI tools can be hugely technical and opaque, making it hard for institutions to explain or properly monitor them for shortcomings. If companies and regulators do not fully understand AI tools, then it is possible they could miss biased or inaccurate results, the report said.
It also noted that AI tools increasingly rely on large external datasets and third-party vendors, which pose their own privacy and cybersecurity risks.
Some regulators including the Securities and Exchange Commission, which sits on the panel, are scrutinizing how firms use AI, while the White House recently issued an executive order aimed at mitigating AI risk.
Elsewhere in the report, the FSOC noted that the U.S. banking system remains resilient, despite large bank failures this year. But it urged regulators to keep a close eye on uninsured bank deposits, the rapid flight of which triggered the failures.