diff --git a/Sources/App/Partials/posts/2026-09-23-welfare.md b/Sources/App/Partials/posts/2026-09-23-welfare.md
new file mode 100644
index 0000000..d1fb478
--- /dev/null
+++ b/Sources/App/Partials/posts/2026-09-23-welfare.md
@@ -0,0 +1,5 @@
+# A Warning About Model Welfare
+
+> An AI model can describe pain in perfect prose without feeling anything, which is the inverse of biological experience. Animals feel first and then describe them later. In LLMs, description is the whole product, and there is nothing that suggests anything is beneath it.
+
+Mustafa Suleyman, the British AI entrepreneur who runs Microsoft AI, [cautions](https://mustafa-suleyman.ai/a-warning-about-model-welfare) against mistaking fluent description for feeling. Fittingly, the warning reads like it could have been written by a model.