Federal Reserve Governor Christopher J. Waller reported on October 1, 2026, that AI agents and bots now generate approximately 50% of the traffic on the FRED economic data platform. This automated activity has contributed to an annual traffic growth rate of about 150% for the service, which currently hosts over 850,000 economic time series.
To accommodate this shift, the St. Louis Fed introduced the FRED MCP Connector. This tool utilizes the Model Context Protocol, enabling users to connect AI agents directly to the database and perform queries using natural language. The organization is also modifying its data documentation and presentation methods to ensure higher accuracy for automated systems.
Despite the integration, Waller highlighted risks regarding data integrity, noting that AI agents may misattribute sources or misinterpret correlations as causation. The Fed is currently addressing these challenges while planning to expand its AI-friendly architecture to include historical archival tools such as FRASER and ALFRED.
