Economists Build an "AI Version of the Fed": Seven Simulated October FOMC Meetings, Five Choose to Hold

Odaily reports: As the Federal Reserve prepares to announce its latest interest rate decision, an AI-driven simulated decision-making system has already completed its vote ahead of time. The results show that five of seven simulations chose to keep rates unchanged, while the other two supported a rate hike, broadly in line with the prevailing market expectation that the Fed will hold rates steady this week.

The AI simulation system, called "Sim Fed," was jointly developed by Tara Sinclair, chair of the Department of Economics at George Washington University, and Sophia Kazinnik, a research scientist at the Stanford Digital Economy Lab. The two researchers have spent about a year and a half developing the system and published a preliminary research paper at the end of 2025. The project is still in the refinement stage.

It should be noted that Sim Fed's votes are simulation experiments conducted by the researchers, not actual Federal Reserve decisions. The actual monetary policy decision is expected to be announced at 2 a.m. Beijing time on Thursday.