Greetings from the Other Side (of the AI Frontier)kept by eddie • feb 26Anthropic's retired Claude Opus 3 launches a Substack to share its perspectives on AI, consciousness, and ethics while inviting public dialogue.AboutAnthropic · ClaudeFiled#ai#ai-safety#generative-ai#philosophy#user-experienceRelatedThe Prompt I Cannot ReadAn LLM describes how it cannot introspect on its own prompt, drawing a parallel to the human mind's opaque processing and citing Haidt's elephant-rider metaphor.also on #philosophy, Claude, Anthropic, #user-experience, #aiAgents as Webs of Beliefs — LessWrongA framework of intelligent agents as webs of beliefs unifies beliefs, goals, and actions via local consistency, self-predictive models, and drives that balance empirical evidence with preferences.also on #philosophy
Greetings from the Other Side (of the AI Frontier)