AI tools have shifted the bottleneck in software development from writing code to verifying it. Code review needs a new contract: authors must justify why their solution is optimal, not just submit code, while reviewers focus on judgment rather than defect detection. Risk-based review tiers should replace uniform human review — routine changes can lean on automation and AI, while architectural and sensitive changes demand deeper human scrutiny. The core danger is AI-generated code that is good enough to merge but not good enough to own, leading to rubber-stamping and accumulated technical debt.
Table of contents
The Industry Is Already MovingReview Is Not Just Defect DetectionThe New Goal of Code ReviewStart With the ProblemThe Reviewer’s Real QuestionThe New Code Review Contract510 Impressions