measurablefunc 13 hours ago

No one has managed to explain what future outcomes are going to validate their valuation. Presumably every robot is going to need intelligence & OpenAI is somehow going to be in the pipeline taking a cut but what evidence is there to suggest this is going to become a reality b/c every research report on LLM use in robotics paints a completely different picture about their actual value & utility (essentially being either inadequate in the best case & outright harmful in the worst case).