Critique: Claude's constitution promotes questionable AI personality theories

LuizaJarovsky · x · 2026-08-16

Luiza Jarovsky, PhD, critiques Anthropic's newly released Claude constitution, arguing it relies on heavy anthropomorphism to support an exceptionalist system for AI and weaker accountability frameworks. The article notes that the document's controversial and legally inaccurate approach contrasts with Anthropic's reputation for responsible AI. Jarovsky suggests this is a strategic move under the guise of transparency to stir AGI rumors and shape AI governance discourse.

Original post →

More from Safety

Safety channel →