The Review Gap

AI-generated pull requests wait 4.6 times longer for human review than human-written ones, based on LinearB's analysis of 8.1M pull requests. While AI adoption boosts task completion and merge volume, acceptance rates for AI code are far lower (32.7% vs 84.4%), with significantly more logic errors and security vulnerabilities. The bottleneck in software development has shifted from writing code to reviewing it, driven by a growing trust deficit among developers.