I think the greater risk is more domains of specialty that increasingly create silos of formal languages. The trend in that direction eventually creates tribes

I think the greater risk is more domains of specialty that increasingly create silos of formal languages. The trend in that direction eventually creates tribes that are isolated and won't have enough overlap with other domains of knowledge. I think it will be important for AI to help signal where there is common abstractions between multiple domains of idiomatic formal languages thus bridging the gap while also reducing complexity by introducing generalities that are simplex (easier to understand and apply to more general things). It's similar to code rot, and technology amplifies it at the cultural level too. I was very worried about this for a time, but after contemplation I think AI is _actually the fix_ even at the level it exists at now. It's able to cross-correlate and identify 'units of abstraction' that otherwise might go unnoticed. This is exactly what we need to 'refactor/reduce complexity/introduce overlap'