My understanding was that GPT4 evaluation appeared to specifically favour text that GPT4 would generate itself (leading to some bias towards gpt-based fine-tunes), although I can't remember the details
Given the possibility of bias, it would make sense to have the judge “recuse” itself from comparisons involving its own output. Between GPT-4, Claude, and soon Gemini Ultra, there should be several strong LLMs to choose from.
I don’t think it would be a replacement for human rating, but it would be interesting to see.