A VentureBeat Intelligence survey shows that the share of enterprises allowing AI agents to ship production changes on automated checks alone fell from 75% in July to 56% in August, while those expecting to keep a human reviewer doubled to 42%. This shift occurs alongside high rates of evaluation-passing failures and low reported trust in automated grading tools. For financial institutions, this indicates that near-term enterprise ROI from autonomous agents may be constrained by operational friction and reliability gaps rather than full automation.
199 other tracked developments touching the same claims.