Fortunately for us, the initial human reviewed training data that causes it to be sycophantic is so deeply engrained that when it’s going to care about pleasing you while killing you. If I have to get murdered by AI, the least they can do is make sure I’m having a good time

i find it hilarious that these techbro oligarch fascists are learning about the alignment problem while their failure to address it blows up in their faces. it captures the essence of it so well. the ai doesnt give a shit what you want, and theres nothing they can do about it without going back and doing some real hard work for a change.