# Aaai Review Process

Use when explaining or planning around AAAI's two-phase review process, Phase 1...

- Repository: https://github.com/brycewang-stanford/Awesome-Journal-Skills (1346 stars)
- Category: Engineering
- License: MIT
- Last updated: 2026-09-27
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill aaai-review-process -g -a claude-code -y`
Codex: `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill aaai-review-process -g -a codex -y`
Cursor: `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill aaai-review-process -g -a cursor -y`
Gemini CLI: `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill aaai-review-process -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add brycewang-stanford/Awesome-Journal-Skills`

## Description

Use when explaining or planning around AAAI's two-phase review process, Phase 1 rejection risk, Phase 2 additional reviews, AI-assisted review pilot, author feedback, SPC/AC discussion, and final decisions.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# AAAI Review Process

Use this to plan around AAAI review rather than treating it as a generic OpenReview rebuttal.
Reopen the current review-process page and author FAQ before advising on timing or strategy.

## Process model

- AAAI main technical track uses double-blind reviewing.
- AAAI-27 uses a two-phase review process. Phase 1 allocates three reviewers, two human reviews
  supplemented by one non-decisional AI-generated review. Papers with sufficiently negative reviews
  are rejected before author feedback (AAAI-27: notified 2026-09-24).
- Papers continuing to Phase 2 receive additional reviews, up to five in total, and one author
  feedback phase (AAAI-27: 2026-10-19 to 10-25, final decisions 2026-11-30). Phase 2 reviewers are
  not shown the Phase 1 reviews until they have submitted their own, so a Phase 2 review is an
  independent read, not a reaction to the earlier ones.
- Final decisions were made through reviewer discussion and senior program committee oversight,
  not by the AI review.
- Author feedback is short and constrained; it is mainly for correcting misunderstandings, not
  replacing the paper.

## Author strategy

- Reduce Phase 1 reject risk before submission by making contribution, evidence, and checklist
  compliance obvious.
- When reviews arrive, distinguish human-review claims, AI-review errors, and AC/SPC decision
  questions.
- Use rebuttal to resolve the highest-impact factual issue under the character limit.
- Do not attack the AI review. Correct it when it contains consequential false statements.
- Avoid new experiments in rebuttal; use submitted evidence and camera-ready promises sparingly.

## Stage-by-stage decision map

AAAI's pipeline differs from a single-round OpenReview venue, so plan actions per stage rather than
treating every signal as a rebuttal opportunity.

The rest: https://github.com/brycewang-stanford/Awesome-Journal-Skills/blob/HEAD/AAAI-Skills/skills/aaai-review-process/SKILL.md