IDK about big companies, but the open source side seems to have realized that heavily safety aligning AI kind of lobotomizes it, yet is still trivial to defeat.
It appears safety tools are better developed as wrappers on a case-by-case basis.
Also, all the noise about emerging AGI is just fearmongering. Near future multimodal LLMs are extremely dangerous, but they are not AGI.
omg I had to read the title 5 times. I kept reading "Why are AI ethnic teams being disbanded" and I'm like wtf?
CEOs and Investors are realising that the interest of AI safety teams mostly aren’t aligned with corporate goals. They were good for press as long as AGI was a distant dream but now as the AI wars start over the big tech companies, there is no need for self-sabotage. The drama at openai was likely the death sentence for serious backing of those teams.