The evolution of answers from version to version makes it clear there are insane amounts of manual fine tunings happening. I think this is largely overlooked by the "its learning" crowd.
What sort of evidence would convince you that it is learning?
To me the LLM loophole/"hack" closings just feel like a human vs human cat&mouse game with some Chat UI in the middle.
Because once it does that thing without you having expressly decided that is the goal, it’s very tempting to just move the goal a liiiitle bit further away
What's more, is we can do that today. Just think of any problem which you suspect won't be included in OpenAI's hand-tunings and check both 3.5 and 4.
They have millions of people training the AI for free basicallly, and they have engineers who pick and rate pieces of training data and use it together with other sources and manual training.