AI agent autonomy remains an elusive work in progress - CIO

AI systems designed to work independently and make decisions without constant human oversight are still unreliable and limited. This matters because companies are deploying these systems in real work while they're not yet ready.