freddy reposted
freddy
freddy reposted
freddy reposted
> me when im eating
I have noticed a trend in food ads where they picture people (mostly women, for reasons that are obscure) taking a big bite of something and doing this kind of "oh, yes, I am definitely eating right now" nodding thing and I just would like all those people to know: I agree with you, you ARE eating.
freddy reposted
freddy reposted
freddy reposted
freddy reposted
What I am getting at is I believe that if there was a credible threat of getting in actual trouble when your AI does something bad that the labs would implement basic simple measures that would be highly effective at preventing those bad things.
Me, personally, I actually believe there are good reasons to think it won’t do these things, and that it is in fact highly controllable. You disagree. Fine. Have it your way. LLMs should be restricted the way nukes are, then.
many are saying bsky.app/profile/jumh...
denaturalization of any sort, for any reason, for anyone, is imo profoundly evil. even for fraud. citizenship should irrevocable. want to stop fraud? find it before you grant citizenship. sizable fine reasonable for citizenship fraud, and other crimes you punish under regular criminal law, sorry!
freddy reposted
freddy reposted
freddy reposted
freddy reposted
also, your periodic reminder that anthropic models do not generate images. anthropic is the only US frontier lab that has been responsible on this one, whether on purpose or not. i will once again ask people to focus on real problems and not make things up just because you're mad at anthropic today.
freddy reposted
freddy reposted