Yes, I agree. Symbolic AI is for hardness, which we can use for the skeleton of the agent. LLMs are the fuzziness that surrounds and interfaces the skeleton. Ev

Yes, I agree. Symbolic AI is for hardness, which we can use for the skeleton of the agent. LLMs are the fuzziness that surrounds and interfaces the skeleton. Everyone who writes some simple code that assembles a promot is already doing that.