Claude "SOUL DOC" reveals something strange...

Anthropic has a new Constitution for Claude. It's like the old "Soul Document" It gives Claude a personality and a set of rules to live by. The reasons behind these documents are absolutely wild... The latest AI News. Learn about LLMs, Gen AI and get ready for the rollout of AGI. Wes Roth covers the latest happenings in the world of OpenAI, Google, Anthropic, NVIDIA and Open Source AI. Claude’s Constitution https://www.anthropic.com/constitution Personality Basins https://near.blog/personality-basins/ The Meaning of Shoggoth AI Memes https://www.lesswrong.com/posts/yjzW7gxk2h7bBs2qr/the-meaning-of-shoggoth-ai-memes The assistant axis: situating and stabilizing the character of large language models https://www.anthropic.com/research/assistant-axis Persona vectors: Monitoring and controlling character traits in language models https://www.anthropic.com/research/persona-vectors Claude 4.5 Opus' Soul Document https://www.lesswrong.com/posts/vpNG99GhbBoLov9og/claude-4-5-opus-soul-document Claude Soul Doc https://gist.github.com/Richard-Weiss/efe157692991535403bd7e7fb20b6695#file-opus_4_5_soul_document_raw-txt Andrej Karpathy Don't think of LLMs as entities but as simulators. https://x.com/karpathy/status/1997731268969304070 ______________________________________________ My Links 🔗 ➡️ Twitter: https://x.com/WesRoth ➡️ AI Newsletter: https://natural20.beehiiv.com/subscribe Want to work with me? Brand, sponsorship & business inquiries: wesroth@smoothmedia.co Check out my AI Podcast where me and Dylan interview AI experts: https://www.youtube.com/playlist?list=PLb1th0f6y4XSKLYenSVDUXFjSHsZTTfhk ______________________________________________ #ai #openai #llm