This blog post discusses two recent research papers from Anthropic that falsely present a known result as a new discovery, highlighting issues in reasoning within the AI field. The author shares personal reflections and critiques the integrity of such findings.