At some point they're going to realise that they need to make cluster of LLM AI's all trained in specialities. Then they will pass "thoughts" to each other, ba

At some point they're going to realise that they need to make cluster of LLM AI's all trained in specialities. Then they will pass "thoughts" to each other, back and forth to self check and verify before outputting. It'll be called something like LPM (Large Philosopher Model). It'll end up chewing up more compute and power than the bloody blockchain, but It'll be really impressive and scare the crap out of Turing testers.