While GPT-4 does very well overall in the comparison rating, there was a consensus among those who DIDN'T prefer GPT-4 responses as to why:
GPT-4 is not preferred when there is “TMI” or “too much information.”
Implication
Striking the balance of “just enough” information - but not too much - may ultimately come down to personal preference…and that will vary of course. Further, techniques of what to present first, and how much to pause, or enable more, are being tested and rolled out as we speak. Will we see a customization in this space whereby audience data COULD be connected to the platforms in order to tailor the STYLE of how the response is presented? A visual person gets more visuals while someone who prefers text gets more text, etc. This already exists in social media and ad tech. It may very well change how content is presented in the future.
The length of the second one (GPT-4) threw me off, it is much too long for just a simple question.
The other response (GPT-4) was too lengthy and provided more details that felt unnecessary.
I thought the second (not GPT-4) one is the most concise answer without any extra feedback that wasn't asked.
The option I did not choose (GPT-4) seemed as if it gave too much information when the asker was only seeking quick, basic information.
The one I didn't choose (GPT-4) felt inhuman, with unnecessary categories before each answer, and didn't explain their reasoning at all. Felt bland.
The second one (GPT-4) sounded more like a robot answer.
