diff --git a/content/pages/ai.md b/content/pages/ai.md
index 4455310..670a227 100644
--- a/content/pages/ai.md
+++ b/content/pages/ai.md
@@ -6,7 +6,7 @@ category:
   - AI
 syndication:
   - https://blog.giersig.eu/ai/
-updated: 2026-03-09T14:54:38.336Z
+updated: 2026-03-09T14:57:12.097Z
 summary: How AI is used on this blog, with structured disclosure levels for text and code.
 visibility: public
 permalink: https://blog.giersig.eu/ai/
@@ -23,9 +23,10 @@ Everything published here is mine: the judgment, the positions, the responsibili
 
 I'm transparent about this because I think the default of silent AI use is dishonest. If something was shaped significantly by AI, you deserve to know.
 
-Thanks to [rmendes](https://rmendes.net/ai/) for this idea!
-
 - Level0 - No involvement at all
-- Level1 - Editorial assistance (eg. grammar correction, minor rephrasing, tone refinement, translation)
-- Level2 - Co-Working (turning bullet points into paragraphs, expanding a draft into a structured essay, generating alternative arguments that I then refine, drafting tutorial text based on my actual codebase, summarizing material I already read.)
-- Level3 - Fully AI generated, human assisted
+- Level1 - Editorial assistance (like grammar correction, minor rephrasing, tone refinement, translation)
+- Level2 - Co-Working (like turning bullet points into paragraphs, expanding a draft into a structured essay, generating alternative arguments that I then refine, drafting tutorial text based on my actual codebase, summarizing material I already read.)
+- Level3 - Fully/Primarily AI generated, human assisted
+
+Thanks to [rmendes](https://rmendes.net/ai/) for this idea!
+