With current generation AI we really need AI + Humans to have good results. It seems likely that the entire LLM branch of products will have this limitation. If that's the case let's race to make the best open source AI as fast as possible, so it can be spread as wide as possible and can be used to fix our shared problems. It's the spreading widely that will lead to breakout results in fields such a cancer research just as much as having the most intelligent system, because humans bring some X factor for creativity/ingenuity/novelty.