Anyone else get the feeling that GPT-3.5 is becoming dumber?
I made an app for myself that can be used to chat with GPT and it also had some extra features that ChatGPT didn't (but now has). I didn't use it (only Bing AI sometimes) for some time and now I wanted to use it again. I had to fix some API stuff because the OpenAI module jumped to 1.0.0, but that didn't affect any prompt (this is important: it's my app, not ChatGPT, so cannot possibly be a prompt cause if I did nothing) and I didn't edit what model it used.
When everything was fixed, I started using it and it was obviously dumber than it was before. It made things up, misspelled the name of a place and other things.
This can be intentional, so people buy ChatGPT Premium and use GPT-4. At least GPT-4 is cheaper from the API and it's not a subscription.
They could've easily continue going the Catalina way (you can allow 32-bit programs to run after a warning if upgraded from an older OS), but they didn't.
I don't understand why they forced 64-bit on Big Sur, it breaks so many old, non-updated apps and they know that.
Ok, we made it so YouTube only fetches an enter URL of a video, and when you click on a video, that opens. Only YouTube can give you the actual video link as it is stored and associated in our own database, and you only get it when logged in with Google (so basically you need an account to watch videos but not to see the thumbnails).
Now, we'll only allow Google account sing-ins from a trusted browser or trusted website. This is of course for your security. We will provide you a custom token if you ask to approve a browser, website or an app and we'll take careful looks to see if it violates our policies or not. If we don't detect a valid token provided by your browser, we will display an error message and won't allow you to sing in. Firefox has to also implement this in order to stay relevant (most people need allowing Google logins). The policy bans all things like scam websites, some other things and things against TOSes including YouTube's. Third-party apps blocking ads will be not permitted by YouTube now. We will reject any approve request done by things like Piped and Newpipe.
Have fun with your broken pipe. If they find a fix, we will break it again (:
Anyone else get the feeling that GPT-3.5 is becoming dumber?
I made an app for myself that can be used to chat with GPT and it also had some extra features that ChatGPT didn't (but now has). I didn't use it (only Bing AI sometimes) for some time and now I wanted to use it again. I had to fix some API stuff because the OpenAI module jumped to 1.0.0, but that didn't affect any prompt (this is important: it's my app, not ChatGPT, so cannot possibly be a prompt cause if I did nothing) and I didn't edit what model it used.
When everything was fixed, I started using it and it was obviously dumber than it was before. It made things up, misspelled the name of a place and other things.
This can be intentional, so people buy ChatGPT Premium and use GPT-4. At least GPT-4 is cheaper from the API and it's not a subscription.