Anthropic believes they are the only people who should control AI.
replyThat alone is terrible. Also OpenAI has done actual work to try and mitigate the downstream negatives effects with UBI experiments.
Sam isn’t perfect but it’s deeply unclear that he’s worse
> Anthropic believes they are the only people who should control AI.
replyI’ve seen this a few times in the thread. Can you or anyone provide a link that supports this claim?
This seems to be a rumor being coordinated by OpenAI.
replyThere's OpenAI employees spreading this rumor on Twitter with 0 evidence. Their entire evidence is "I keep hearing Anthropic wants to control AI". Their evidence is literal rumors.
I wouldn't say Anthropic is worse than OpenAI, but there's a lot wrong with them. https://anthropic.ml/ has a collection of incidents and relevant evidence.
replyIsn’t that site just loudly proclaiming that “pushing the frontier” of AI is inherently anti-social and that employees should be asking the public to “shut us all down” ?
replyThat’s their core argument against Anthropic, that they are making progress at improving their models ?