Copilot code review auto-resolves comments, writes commit messages

Sept 28 also brings a policy overhaul: chat data kept for your account's lifetime, review default flips to Balanced — and Codex adds ultra reasoning tiers.

Nowline SEP 13 1:00 PM banner

Top AI stories from the last hour

Top AI stories from the last hour

Copy markdown

  • Reviews that clean up after themselves

    Copilot code review now auto-resolves its own comments once you address them and writes smart commit messages for you — less bookkeeping per review loop. Live now on GitHub.

  • Codex unlocks “max” and “ultra” reasoning

    OpenAI's Codex 0.154.0 adds max and ultra reasoning-effort tiers for your hardest tasks, plus richer resume/fork with include_turns. Heads-up: ultra can backfire on routine work — bigger budgets mean slower, pricier runs.

  • Heads-up: Copilot's Sept 28 policy merge

    On Sept 28, Copilot Chat across github.com, Mobile, and the cloud agent unifies under one policy, and chat data retention jumps from 28 days to your account's lifetime. You can opt out, but it removes Copilot on github.com and Mobile.

  • Set your review effort before Sept 28

    Copilot code review's default flips from Lite to Balanced on Sept 28. If you rely on Lite for speed or cost, change your repo or org settings before then to keep it.

  • New Copilot seats bill upfront Oct 1

    New Copilot Business and Enterprise seat assignments now require payment before access; existing customers see it from Oct 1. Budget for per-seat charges before adding teammates.