Hey all, just uploaded a new video! Andrej Karpathy (OpenAI's original co-founder and former head of AI at Tesla) just open-sourced the LLM Council - get ChatGPT, Claude, Gemini, and Grok to answer the same question, debate each other's responses, rank who gave the best answer, and have a fifth AI synthesize everything into one final battle-tested response. The Problem: You ask ChatGPT a question, it sounds confident, so you run with it. But every model has blind spots, biases, and will confidently give you wrong information sometimes. When making important decisions, you want multiple perspectives that challenge each other - just like a real board of directors. What This Does: Forces 4 different AI models to answer independently, then see each other's responses (anonymized), critique and rank each other, and a chairman model synthesizes the final answer. You get a response that's been pressure-tested by 4 different AI systems before you act on it. Business Use Cases: Validate expensive ideas before dropping $15K on consultants or $50K on campaigns. Review contracts and proposals where AI models ruthlessly point out unfavorable terms. Fix problems with distinct theories instead of getting stuck on the first thing that sounds right. Make big decisions (expanding, hiring, launching) with battle-tested plans instead of single AI guesses. Watch here: https://youtu.be/6zsLHbvUPQM?si=YZkNgnwdUSd-Y_Ne https://github.com/karpathy/llm-council.git