I doubt arguing about whether LLM are "AGI" or have "agency" is useful to developers at all. But I also don't understand what the usefulness is of claiming arc

I doubt arguing about whether LLM are "AGI" or have "agency" is useful to developers at all. But I also don't understand what the usefulness is of claiming architectures that are entirely passive- only speaking when spoken to, have "agency." If you want a definition, let's start with a neural network that runs ongoingly, does some sort of goal driven, useful behavior, and doesn't merely respond to prompts. Here's a different more sober take on current models: https://twitter.com/ylecun/status/1718263303485501784?s=19 https://twitter.com/ylecun/status/1718263303485501784?s=19