@piotr_bartoszek, I agree that it’s great for research work, and I rely on it for that purpose as well. But, I find ChatGPT and Claude to be better for generating content.
ChatGPT have the feeling that they are really on the forefront, with every new update. Also 4o and the new research update are really good additions. Sometimes I have the feeling that AI does make mistakes for example in coding, and heard that the new research version makes less mistakes.
@wouter_van_der_meij I’ve heard great things about the research version too, but I haven’t tested it out yet. I want to test how it compares to Perplexity AI in terms of accuracy and performance.
@wouter_van_der_meij@rupal_saini Perplexity isn't perfect either... I find it say the same thing as the question, without fact checking. It's almost like a bug. It works sometimes, and it doesn't at other times. And this is Dec 2024 update.
These days, I mainly use Claude, and when I'm not satisfied with Claude's responses, I use GPT.
Claude's writing ability is excellent. It's incredibly good at Korean.
Also, when asking coding-related questions, Claude's artifacts feature, which GPT doesn't have, is extremely convenient.
@nicolepark, thanks for sharing your experience! I’ve also been using Claude for the past week and I’m pretty impressed by its writing ability—it's truly top-notch.
Owwly