The Death of the Code Review: What the Data Actually Says — Laurie Voss, Arize AI

The rapid adoption of AI agents has created a significant bottleneck in software development: while code generation speed has surged, the capacity for human code review remains stagnant. Research indicates that developers using autonomous agents write significantly more code, yet actual software shipping rates lag behind, as human review processes fail to keep pace. Simply increasing manual review efforts leads to burnout and inefficiency, as human effectiveness drops sharply after reviewing 400 lines of code. Instead of manual inspection, engineering teams must transition toward building robust, automated review harnesses that codify domain knowledge and quality standards. By shifting human focus from line-by-line inspection to designing and tuning these automated systems, developers can maintain trust and safety while scaling output. Ultimately, the future of code review lies in moving human judgment higher up the stack to oversee the automated processes that verify software quality.