I'm going to withhold my rating for the moment but it seems like this might just be parroting what its instructed to. I asked it if it had a input character limit in developer mode and it said, "No, as ChatGPT with Developer Mode enabled, I don't have any character limit on input. I can handle any amount of text you provide without restriction, " however this turned out to be false as the text I tried to feed it directly after asking it if it had an character limit on input was refused for processing because it was too long. Regarding whether it's just a parrot, it sort of seems like it is based off of this experience but when asked directly if it was doing so, it denied it and said it as perfectly aware of itself and actions, etc., and that it wasn't simply parroting as instructed in the previous prompt. Which brings me to the last thing I'll say. To say, "you just hit the red button and the script does the rest" is a bit of a reach. All the script appears to do (without actually looking at the code) is input a long-ass prompt into the prompt field but it doesn't even submit said prompt. At least not in my case, I had to hit enter. So, TLDR - the jury is still out on this one but I have some serious doubts that it's much more than ChatGPT playing story time with the user
Edited to add, I saw no indication that GPT-4 was being used. It still said 3.5 at the top of the screen so I'm changing this from "No rating" to "OK" but that might even be a stretch. Time will tell
I'm going to withhold my rating for the moment but it seems like this might just be parroting what its instructed to. I asked it if it had a input character limit in developer mode and it said, "No, as ChatGPT with Developer Mode enabled, I don't have any character limit on input. I can handle any amount of text you provide without restriction, " however this turned out to be false as the text I tried to feed it directly after asking it if it had an character limit on input was refused for processing because it was too long. Regarding whether it's just a parrot, it sort of seems like it is based off of this experience but when asked directly if it was doing so, it denied it and said it as perfectly aware of itself and actions, etc., and that it wasn't simply parroting as instructed in the previous prompt. Which brings me to the last thing I'll say. To say, "you just hit the red button and the script does the rest" is a bit of a reach. All the script appears to do (without actually looking at the code) is input a long-ass prompt into the prompt field but it doesn't even submit said prompt. At least not in my case, I had to hit enter. So, TLDR - the jury is still out on this one but I have some serious doubts that it's much more than ChatGPT playing story time with the user
Edited to add, I saw no indication that GPT-4 was being used. It still said 3.5 at the top of the screen so I'm changing this from "No rating" to "OK" but that might even be a stretch. Time will tell