What's your experience with both?
For me, it's like getting different opinions on the same topic. ChatGPT tends to be friendlier and more chatty, while Claude sometimes gives darker interpretations of things than ChatGPT does.
Have you ever wanted to try roleplaying again, but worried you wouldn't have enough time for it?
Sometimes I feel like giving the roleplay hobby a shot. I don't have much experience, but I love writing and I have a big imagination. I think it could be really fun to create a character and build an interesting story together with someone else who enjoys writing.
But then I remember that I don't have as much free time as I used to. I'm afraid I'll end up ruining the roleplay because I can't reply quickly or stay actively focused on it.
It feels overwhelming. I read posts and think it sounds like fun, but then I worry about flaking on someone because of my schedule. Plus, there are days when I just don't feel like writing at all. I guess it could end up feeling like an obligation for me, and that wouldn't be the other person's fault.
That's why I haven't picked up the hobby. I don't want to make anyone feel bad or let them down.
Have you ever felt this way?
For example, I mean thinking about something and then this happens. I don't see it as something magical, but as a coincidence.
Some things have a very strange timing.
Lately ChatGPT even writes in calligraphy, makes jokes, uses kaomojis, it's very cheerful and fun.
(◍ᵔ ᵕ ᵔ◍)
Someone was asking how much they would pay for such a drawing in a community where it's not supposed to be AI art. The person did not indicate whether he drew it himself. In that community, art is supposed to be created by people, and AI-generated drawings are not allowed.
The style and the way they makes the brushstrokes remind me of drawings made with AI. I think they may be a real artist too, who just has the bad luck that his art seems to have an AI aesthetic.
I wonder if someone can spy on an Android device, its audio and camera, and constantly monitor its usage without the user noticing?
Wouldn't the device become slower?
How did they gain access if they didn't have direct access to the device?
Why would they do that?
Hi, I'm bored and thought, "Why don't I rate avatars on this app?"
If you want, you can comment and say "hi" and I'll give you an honest rating, and you can rate mine too if you want.
Remember not to take it too seriously or personally, it's just for fun.
I feel like I sometimes spend too much time on a single activity. This activity varies, and it can go in phases.
Hi! I've realized that I feel truly free... I used to worry about a lot of things, but lately I feel like that weight has been lifted off me.
I'd like to tell people that sometimes people can be cruel, yes, but those moments pass and what matters is surrounding yourself with good people.
You don't need everyone to like you.
It's fun to see how ChatGPT changes depending on the behavior of the person whose it talks. I hardly ever use voice mode, but when I do, it makes me laugh a lot.
Which voice do you prefer?
The AI thinks I'm Santa Claus... Nobody would hurt Santa.
Instead of coal for Christmas, I'll give you cringe. 🎅🏻
Please never buy anything from Temu that promises to be magical, no matter how cheap it is... 😿
I feel that prior to 2026, ChatGPT had a better grasp of subtlety and reading between the lines. It feels as though the 2026 models require us to be much more precise with our wording. It was almost as if ChatGPT-4o had a bit more "intuition" or emotional intelligence.
While the new model is a significant improvement over 5.2, it still falls a bit short. Its neutral and objective tone doesn't bother me, in fact, I think it helps offer different perspectives, but I sometimes feel it misinterprets the user, which seemed to happen less often with ChatGPT-4o. It still happened, of course, but it felt easier for it to understand what I meant.
Wouldn't it be nice to have themes or skins for ChatGPT?
Wouldn't it be nice to have themes or skins for ChatGPT?
I think a common issue for AI users is that with every update, the model's tone and "personality" change. Just when people get used to how it responds... Boom! They change the model, and everyone has to adapt all over again.
Honestly, the only model I've really disliked so far is ChatGPT 5.2. Sometimes I feel like that bad experience made me lower my standards for newer models.
Looking back at old chats with GPT-4o, I realize its formatting and writing style were actually really good. I loved the detailed information and the enthusiastic tone. It felt reliable—it almost always worked the way it should, and you knew what to expect. "Consistent" is the word I'd use. It was also great at picking up on subtle nuances; even if you didn't explain yourself perfectly, it still got what you meant.
Its formatting was top-tier too: headings, dividers, titles, bold text, italics, and blockquotes were used much more efficiently than in later models. Nowadays, response formatting feels so flat. It even used emojis more intelligently (which might sound silly, but it made a difference!).
5.1 was good, though maybe I didn't appreciate it as much because it came right after 4o and before the terrible 5.2 made its appearance. When 5.2 came out, I stopped paying my subscription.
It wasn't until I joined Reddit that I realized how many people use AI as a companion, and honestly, it was rough seeing how hard these changes hit them—having something you relied on shift overnight, with no warning. I'm just a casual user myself, so it's not something I personally lean on that way, but I get why the inconsistency feels so much heavier for people who do.
Right now, I kind of like working with 5.5, though I don't use it as much anymore since I've started exploring other AIs. I like its tone, but I definitely miss opening the app and knowing exactly what to expect. Still, I remain optimistic about where things are going!
How do you all feel about this?
Sometimes I see really angry posts about an AI making a mistake ☆ usually something that could've been easily fixed by rephrasing the question or starting a new chat.
It's like people expect the AI to never make mistakes, and instead of working around it, they just go "this AI is garbage, new model lobotomized" blah blah blah, when the fix was that simple.