antifuchs

joined 2 years ago
[–] antifuchs@awful.systems 4 points 2 days ago (1 children)

It’s no comfort that an acquaintance who works at a slop company says they’d been doing that for years now. That surely makes it fine then.

[–] antifuchs@awful.systems 9 points 3 days ago (3 children)

I’ve never seen a more compelling reason to enforce strict and short retention rules for every corp communication medium, holy shit https://www.forbes.com/sites/annatong/2026/04/16/ais-new-training-data-your-old-work-slacks-and-emails/

[–] antifuchs@awful.systems 7 points 6 days ago

Excited for the labor and contract law disputes that this will spawn when the model makes promises that the person won’t keep https://www.theverge.com/tech/910990/meta-ceo-mark-zuckerberg-ai-clone

But as as expected, this is another zuck project that doesn’t have the leg(itimation)s

[–] antifuchs@awful.systems 5 points 6 days ago

[ai booster voice] if you last tried a head of AI more than six months ago, you need to try the new model. https://old.reddit.com/r/apple/comments/1ska7kn/apples_ai_chief_john_giannandrea_departs_this_week/

[–] antifuchs@awful.systems 10 points 1 week ago

I hate knowing so many of these clowns in person (or thinking I know them, Jesus fucking Christ) https://futurism.com/artificial-intelligence/openai-staffers-horrified-insane-plan

[–] antifuchs@awful.systems 9 points 1 week ago (7 children)

Is it their next model that tbey swear isn’t vaporware but no! It is too dangerous to release into the world because it’ll find too much insecure code or whatever.

[–] antifuchs@awful.systems 11 points 1 week ago

Aphyr weighing in with an ai position post:

Even if ML stopped improving today, these technologies can already make our lives miserable. Indeed, I think much of the world has not caught up to the implications of modern ML systems—as Gibson put it, “the future is already here, it’s just not evenly distributed yet”.

[–] antifuchs@awful.systems 9 points 2 weeks ago

For my own services I’m aiming for .999999% of uptime

[–] antifuchs@awful.systems 17 points 2 weeks ago

You just gotta appreciate that the most deterministic part of the code base is the predatory lootbox system.

[–] antifuchs@awful.systems 12 points 2 weeks ago (8 children)

Hold on now, the uptime number contains two digits that are nines! The image itself has four nines in total!

[–] antifuchs@awful.systems 6 points 2 weeks ago (5 children)

This thread by Johnny reading (skimming on a phone, hah) through it is really good.

If only literally any human with context and a small screen to look at the bigger picture was involved with decisions around taking this to production, it would … still be bad but only on a societal level.

 

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.

view more: next ›