I always double check with Google searches, but at least ChatGPT gets me somewhere where I can actually search for something useful.
Relatedly, sometimes, even with an initial prompt I've used in the past to make it do what I need to a text in Japanese, and despite everything I type being in Japanese, sometimes it decides to respond entirely in English.
For example I used it over like 30 minutes to conceptualize, solve, and write some code that draws a graphic for a simple physics problem (0-shot) that I could intuitively understand but had no (physics and math knowledge) tools to calculate properly and it was a great experience.
It let me pick and change how what properties i wanted represented on the graph, knew what center of mass means, how to calculate it for a weird object, got something that feels correct to me, drew it out with various representations where i used color, size, shape to represent the various distances, weights, clusters, etc.
My non programmer friends have used generative networks to make designs for prints, incredible and generally accurate folk art, full mobile games.
People are sleeping on this. Don't be them.
That the model is better in English is no surprise given that most of its training corpus is in English. In fact, based on the sentence structure of the output when it speaks in Russian, it's clear that what's happening there is some kind of real-time translation from English.
That aside, I have yet to see any task on which GPT-4 wasn't at least as good as, or better than, GPT-3.5. I'd love to experiment with that. Do you recall any specific examples?
For example, I am studying Japanese and often encounter expressions that seems to mean the same thing; I would ask my teacher, but her time is limited. I can instead ask ChatGPT and only bring to my teacher the questions whose ChatGPT answers did not convince me.
Another example: I like understanding why there are certain steps in a recipe. It would be hard to find someone with the knowledge and time to answer the question, never mind I should pay for their time. ChatGPT can explain what I want to know at the level of detail that I want.
I was also able to get a decent understanding of a mathematical question I had no business understanding by recursively asking questions until I was able to link its answers to my own knowledge.
It was also able to answer questions about the Spring framework that I had while reading the documentation itself. In that context, going in rabbit holes severely slows down learning and has the potential to just get me lost.