antifuchs

joined 2 years ago
[–] antifuchs@awful.systems 15 points 1 day ago

Oh here’s a fun scenario: FAANG stocks are correcting down and now there’s concern Anthropic will have timed their IPO perfectly such that the investors make out great and the employees have to deal with very much less liquid stock than when it triggered (leading to a massive tax bill they might not be able to cover)

So…

Roko’s Taxilisk.

[–] antifuchs@awful.systems 8 points 3 days ago (1 children)

Jan has been putting out some good posts lately and I felt the latest one in my bones

[–] antifuchs@awful.systems 6 points 4 days ago

Is it bad that my first thought was “at least she isn’t bragging about hitting her kids”?

Anyway, tech parents stop telling on yourselves. Challenge level impossible

[–] antifuchs@awful.systems 10 points 5 days ago

Why would anyone get excited about EA now that they’ve killed beloved games studio BioWare

[–] antifuchs@awful.systems 6 points 2 weeks ago (7 children)

Ah yes, Nazis, well known for speaking english

[–] antifuchs@awful.systems 1 points 2 weeks ago

Brand recognition

[–] antifuchs@awful.systems 3 points 2 weeks ago (1 children)

Weird, it’s a regular Google trust services issued cert here; my Firefox doesn’t complain about it.

[–] antifuchs@awful.systems 18 points 2 weeks ago (9 children)

Feel like this blog must have made the rounds but today is the first time I’m actually reading it and it’s sharp enough to cut with:

Every death on my corner is offset, with a third of a toddler left over, and I’d like that number in the conversation the next time somebody calls me a murderer out the window of a passing Kia.

[–] antifuchs@awful.systems 13 points 3 weeks ago

Translation: they have started enforcing request rate limits.

[–] antifuchs@awful.systems 9 points 1 month ago

One more: I gotta check with my physical therapist if her job is safe now that RSI is being taken over by AI

[–] antifuchs@awful.systems 9 points 1 month ago

Jokes on them, I’ve developed chronic RSI for free, decades ago

 

Got the pointer to this from Allison Parrish who says it better than I could:

it's a very compelling paper, with a super clever methodology, and (i'm paraphrasing/extrapolating) shows that "alignment" strategies like RLHF only work to ensure that it never seems like a white person is saying something overtly racist, rather than addressing the actual prejudice baked into the model.

 

School student tells AI to put 20 other students’ faces on nude pictures, shares them in chat; it takes months for anyone including the school administrators to act because of some extremely, uh, dubious loophole.

If someone does that in photoshop, it’s a crime; if they do it in AI pretending to be photoshop, it’s somehow not. Gotta love this legal system’s focus on minor technicalities rather than the harm done.

 

They have Nik Suresh (the author) on, as well as Robert Evans. I haven’t listened to it all yet, but it’s fun so far.

view more: next ›