In a speech delivered at a National Bureau of Economic Research (NBER) event in Cambridge, Massachusetts, Cook said the Federal Open Market Committee (FOMC) is not using AI in developing policy, but is using it to help with writing, coding and research.

“For example, we have been deepening our understanding of the capabilities of LLMs [large language models] and other machine learning models to produce economic insights,” Cook said, adding that what they learn is being documented in Fed papers.

In addition to learning about AI tools by using them, Fed staff regularly engage with professionals from academia, other central banks and industry to keep up with the latest developments, Cook said.

They are proceeding with caution, because they have also encountered the limitations of the technology. For example, Cook said, “When it comes to real-time analysis, LLMs suffer from look-ahead bias and frequently get confused by the vintage nature of economic data releases.”

Offering some guiding principles for adopting AI, Cook suggested that organizations establish strong governance and risk management, educate and train staff, encourage teams to get hands-on experience using AI within controlled environments, and experiment.

Cook also addressed AI’s implications for monetary policy, saying that the technology could affect both sides of the Fed’s dual mandate: maximum employment and price stability.

“As with other technological innovations, AI is poised to reshape our labor market, which in turn could affect our notion of maximum employment or our estimate of the natural rate of unemployment,” Cook said.

She said the technology is likely to make workers more productive, change the tasks associated with jobs and eliminate some jobs.

As for price stability, Cook said, AI could reduce inflationary pressures by increasing productivity.

She added “It is possible that the disinflationary effect of AI could, over time, counter any factors putting upward pressure on inflation. It is also possible that AI could boost prices in the interim, as adoption of the technology might require a surge in aggregate investment.”