AI Agent Reliability Tracker

> recent capability gains have yielded only small improvements in reliability.

Have I missed something? Why would one expect capability gain to make any such improvement?