AI agents still struggle with multi-DB tasks
A new benchmark reveals that AI agents fail badly when they must find, coordinate, and query multiple databases at once.
A new benchmark reveals that AI agents fail badly when they must find, coordinate, and query multiple databases at once.