I give you a lot of credit for your communication and openness. I am afraid full-autonomy is a fairy tale to rationalize the ungodly expense of LLMs, I personally would find taking money on spec for something like fully autonomous robotics to be a little shaky, ethically speaking. But then, I don't believe this is an unsolved problem, I believe this is an unsolvable problem, so more of a philosophical difference than an empirical position at this point.