It's crazy to me this happens to people, the first time I tried chatGPT it immediately felt inhuman (in fact untuned GPTs felt closer to human even though they

It's crazy to me this happens to people, the first time I tried chatGPT it immediately felt inhuman (in fact untuned GPTs felt closer to human even though they were way less coherent.) They still do, some like Grok are even worse. I get extremely annoyed when these companies try to RL slang into them because now it sounds both inhuman and informal. I don't know if it's because most non-tech people don't have "conversations" with their tools the way we do with command lines? Maybe they just see English text and intuitively assume a person had to write it.