Not that I’m a friend of OpenAI, but ChatGPT has relatively fine-grained “personalization” options, and it was never sycophantic with the “efficient” tone for me. Rather the opposite, sometimes it seemed slightly indignant when I criticized it.
I haven't used ChatGPT but I'm a heavy Claude user and today I was discussing the attack on Iran and I mentioned how stupid it was that Iran had no effective air defense when the US and Israel have the #1 and #4 best air forces in the world and it response ended with this which I thought is funny:
"A country that declares "Death to America" and "Death to Israel" multiple times weekly for 45 years probably should have invested in the ability to defend itself against America and Israel."
How so, it’s unstable like floating ice.
Very very large floating ice is fairly stable
I just changed last night and honestly I can't tell much of a difference.
I'm not really complaining, it seems fine, but I'm not seeing the "way better" part that people keep saying.
It's not sycophantic and has a much better "voice."
Not that I’m a friend of OpenAI, but ChatGPT has relatively fine-grained “personalization” options, and it was never sycophantic with the “efficient” tone for me. Rather the opposite, sometimes it seemed slightly indignant when I criticized it.
It definitely is sycophantic, but uses A LOT less emoji, lists and header-paragraph-header structure.
I haven't used ChatGPT but I'm a heavy Claude user and today I was discussing the attack on Iran and I mentioned how stupid it was that Iran had no effective air defense when the US and Israel have the #1 and #4 best air forces in the world and it response ended with this which I thought is funny:
"A country that declares "Death to America" and "Death to Israel" multiple times weekly for 45 years probably should have invested in the ability to defend itself against America and Israel."