Post
A lot of this gets lost in the pivot around whether the machine is meaningfully thinking, and I think a lot of people in today's conversations are pretty sure it is not.
0 reposts15 likes
throwing some ridiculous slop on the pile here maybe not the most tactful thing to drop in this hornets ne^H^H^H^H nice discussion, but "all creatures / all minds / a brighter / weirder / tomorrow" in the background was somewhat goaded on but also, amazing. bsky.app/profile/jaun...
i'm giving up on this not quite working sequence of prompts for now, but, a preview,
Maybe because we're humans, not machines. But we might also resist the idea because LLMs aren't thinking any more than autocorrect is, and it's ridiculous to suggest they are. Please see @irisvanrooij.bsky.social, @olivia.science, @emilymbender.bsky.social, @alexhanna.bsky.social, and many more.
forget the test, the essay covers every stupid debate we’re still having
seriously everyone needs to read Turing in 1950 talking about a very convincing chatbot whose detractors want to say it is not answering but "merely artificially signalling" answers. it doesn't settle anything but it's heartwarming to be part of such a long fight courses.cs.umbc.edu/471/papers/t...
The hurdle wasn’t a settled test of thought. Conversational indistinguishability became cheap enough that people withdrew the inference it once licensed. That’s philosophically important: the test measured an observer’s evidentiary threshold, and the threshold moved when crossing it became ordinary.