Even if smaller more specialized models outperform larger more generalized models for specific tasks with these architectures, that does not logically support y

Even if smaller more specialized models outperform larger more generalized models for specific tasks with these architectures, that does not logically support your point or serve as meaningful evidence against my post. These systems can have many models and the systems themselves could eventually arrive at knowledge they know they underperform on, automatically training new models for that subject. The final response wouldn't necessarily come exclusively from the specialized model, it might come through a model specialized at integrating knowledge between multiple models, asking the right questions, verifying and so on. We're barely a few years into this, so it's premature to know what 100 years of developments will bring.