AI code review
AI code review is the use of large language models to review pull requests automatically — flagging bugs, suggesting improvements, checking for security issues, enforcing style. The dominant tools (GitHub Copilot Reviews, Cursor, CodeRabbit, Greptile, Stride) emit per-line comments on the diff and frequently a summary of the PR's intent.
The practical model is augmentation, not replacement: AI catches the routine review feedback (obvious bugs, missing error handling, style inconsistencies) so human reviewers can focus on the harder judgement work (is this the right design? does it fit the codebase's conventions? will it cause production issues?). Empirical evidence (Microsoft Research 2024, GitHub Octoverse 2025) suggests AI review reduces median review time by 30-50% on routine PRs without measurable defect-rate increase. The risk is that teams stop reviewing AI-generated code as carefully — the appearance of review (AI signed off) substitutes for actual review. Healthy norm: AI review is a layer, not a replacement, and at least one human reviewer must approve.
Related terms
- Code review
Code review is the practice of having another engineer evaluate proposed changes before they merge.
- AI pair programming
AI pair programming is the practice of working alongside an AI coding assistant (Claude, Copilot, Cursor, Continue) as a continuous collaborator on coding tasks — suggesting completions, generating tests, explaining unfamiliar code, drafting refactors.