You can also prompt it to hold back if it doesn’t know, which seems to make a difference. It’s part of my default prompt, and since I added it I haven’t had any overt hallucinations. Definitely invalid code, but not due to crazy errors. Just syntax and inconsistent naming mostly.

I verify just about everything that I ask it, so it isn’t just a general sense of improvement.

Why would anyone pay for something if the free trial doesn’t work? “Hey, you know how we gave you a product that doesn’t quit work as you expect and is super frustrating? Just pay us money, and we’ll give you the same product, but it just works. Just trust us!”

GPT-4 is not the same product. I know it seems like it due to the way they position 3.5 and 4 on the same page, but they are really quite separate things. When I signed up for ChatGPT plus I didn't even bother using 3.5 because I knew it would be inferior. I still have only used it a handful of times. GPT-4 is just so much farther ahead that using 3.5 is just a waste of time.

Would you mind sharing some threads where you thought ChatGPT was useful? These discussions always feel like I’m living on a different planet with a different implementation of large language models than others who claim they’re great. The problems I run into seem to stem from the fundamental nature of this class of products.

The usefulness of ChatGPT is a bit situational, in my experience. But in the right situations it can be pretty powerful.

Take a look at https://chat.openai.com/share/41bdb053-facd-448b-b446-1ba1f1... for example.

A great example. Here's a similar one from me: https://cloud.typingmind.com/share/d2000ffc-a1bf-4b71-b59d-c....

Context: had a bunch of photos and videos I wanted to share with a colleague, without uploading them to any cloud. I asked GPT-4 to write me a trivial single-page gallery that doesn't look like crap, feeding it the output of `ls -l` on the media directory, got it on first shot, copy-pasted and uploaded the whole bundle to a personal server - all in few minutes. It took maybe 15 minutes from the idea of doing it first occurring to me, to a private link I could share.

I have plenty more of those touching C++, Emacs Lisp, Python, generating vCARD and iCalendar files out of blobs of hastily-retyped or copy-pasted text, etc. The common thread here is: one-off, ad-hoc requests, usually underspecified. GPT-4 is quite good at being a fully generic tool for one-off jobs. This is something that never existed before, except in form of delegating a task to another human.

I agree that none of the problems people have mentioned above happen with GPT4.

It used to be more reliable when web browsing worked, but it's still pretty reliable.

Here's a convo I had yesterday when thinking about how to print a Binary Search Tree.

https://chat.openai.com/share/338e7397-0201-44f4-a2c3-75b733...

I use ChatGPT for all sorts of things - looking into visas for countries, coding, reverse engineering companies from job descriptions, brainstorming etc etc.

It saves a lot of time and gives way more value than what you pay for it.