Update Docs about AI policy enforcement

getting-started/ai-tools.rst

@@ -62,3 +62,18 @@ disrespectful of the maintainers time.
 It is not acceptable to alter or bypass existing tests, or remove desired
 functionality, in order to make a failing test pass. Such changes are not
 a real fix.
+
+How these guidelines are applied
+================================
+
+Using AI tools is not a reason on its own for an issue or PR to be labeled
+or closed. When a maintainer thinks an issue or PR does not follow these
+guidelines, they may apply one of the following labels. A bot then posts a
+comment that links to this page.
+
+- :gh-label:`AI-needs-explanation`: the issue or PR stays open. Reply in your
+  own words: explain what the change does or what the issue is about, and
+  describe how you checked that it is correct.
+- :gh-label:`AI-guidelines-not-met`: the issue or PR is closed. If you think
+  we misunderstood your contribution, reply on the issue or PR. You are
+  welcome to explain your approach or ask questions about these guidelines.

triage/labels.rst

@@ -120,10 +120,12 @@ Other labels
   AI-assisted submission where you think the author is unlikely to engage with
   feedback. The bot links to the :ref:`guidelines for using AI tools
   <ai-tools>` and closes the issue or PR. See the :ref:`triage team's guidance
-  on closing PRs <triage-team>`.
+  on applying the AI policy labels <triage-ai-labels>`.
 * Use :gh-label:`AI-needs-explanation` when you have concerns
   about AI-assisted work and think the author is likely to engage with feedback.
-  The bot posts a reminder and leaves the issue or PR open.
+  The bot posts a reminder and leaves the issue or PR open. See the
+  :ref:`triage team's guidance on applying the AI policy labels
+  <triage-ai-labels>`.
 * :gh-label:`build`/:gh-label:`performance`: for issues related
   to the build process or performance, respectively.
 * :gh-label:`easy`: for issues that are considered easy.

triage/triage-team.rst

@@ -56,6 +56,38 @@ the Python Software Foundation, always follow the `PSF Code of Conduct`_.
 .. _PSF Code of Conduct: https://policies.python.org/python.org/code-of-conduct/
 
 
+.. _triage-ai-labels:
+
+Applying the AI policy labels
+=============================
+
+Use the AI policy labels for issues and PRs that don't follow the
+:ref:`guidelines for using AI tools <ai-tools>`. Don't apply them only because
+AI tools were used, or because the author disclosed using them.
+
+Apply only one of these labels to an issue or PR. If you add one when the
+other is already there, remove the other label.
+
+* Start with :gh-label:`AI-needs-explanation`. The bot asks the author to
+  explain the work in their own words and leaves the issue or PR open.
+* Use :gh-label:`AI-guidelines-not-met` only for clear cases, such as a
+  low-effort submission where the author is unlikely to engage with feedback.
+  The bot posts a comment and closes the issue or PR. If you have any doubt,
+  use :gh-label:`AI-needs-explanation` instead, or consult a core developer.
+* If the author doesn't respond to :gh-label:`AI-needs-explanation`, or the
+  response doesn't address the concern (for example, it is more generated
+  text), remove :gh-label:`AI-needs-explanation` and add
+  :gh-label:`AI-guidelines-not-met`. The bot posts the closing
+  comment and closes the issue or PR. If you have any doubt, consult a core
+  team member first.
+* If the author responds and the concern is resolved, remove the label.
+  For a closed issue or PR, reopen it as well.
+
+Prefer these labels over closing manually, so the author gets the same
+explanation each time and the issues and PRs can be found later by filtering
+on the label.
+
+
 Becoming a member of the Python triage team
 ===========================================
 

triage/triaging.rst

@@ -28,6 +28,8 @@ Checklist for triaging
 * If the issue is clearly invalid (unrelated to CPython, spam, and
   so on), you can use GitHub's :guilabel:`Close as not planned` option
   and apply the :gh-label:`invalid` label.
+* If the issue or PR doesn't follow the :ref:`guidelines for using AI tools
+  <ai-tools>`, apply one of the :ref:`AI policy labels <triage-ai-labels>`.
 
 
 .. _triage-fields: