avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

www.greaterwrong.com/posts/yFTMGK... rationalists are fuckin geniuses*

aug 30, 2025, 1:13 pm • 39 6

Replies

avatar
Christian Paulsen @christianspaulsen.bsky.social

The claim that an AI translation of an article will be just as simple and accurate as that. A "1:1 translation" misunderstands translation. Also, "AI" hardly "speaks all languages". Leaving aside unwritten languages, it depends on sufficient corpus to have anywhere a decent chance of making sense.

The first idea is translation to languages other than English. Those languages often have fewer speakers, and consequently fewer Wikipedia volunteers. But for AI encyclopedia, volunteers are not a bottleneck. The easiest thing it could do is a 1:1 translation from the English version. But it could also add sources written in the other language, optimize the article for a different audience, etc. We can further expand on these ideas. The AI speaks all languages, so in principle, it could use sources from all languages in all language versions. The problem is that the human reader wouldn’t understand most of them, so the references would be useless for them. However, different people speak different sets of languages, so maybe we could let the reader specify which languages they are comfortable with, display references to sources written in those languages, and hide the rest? So when you read an English article on e.g. Eiffel Tower, by default you only get references to English sources, but if you specify that you also speak French, you also get references to French sources. That could provide extra value if e.g. some detail is not mentioned in the English sources.
aug 30, 2025, 2:15 pm • 2 0 • view
avatar
Christian Paulsen @christianspaulsen.bsky.social

Likely, you end up with something that appears (for the uninformed) as accurate, but for any but the most represent languages online, is likely riddled with poor grammar or inappropriate Anglicized structure or vocabulary. At least the "Scots" wiki hoax articles had an intentionality behind them

aug 30, 2025, 2:36 pm • 2 0 • view
avatar
Soyweiser @soyweiser.bsky.social

I know there is a running joke about people starting a new project which will go nowhere and somebody says 'ill make the wiki' but this isn't even making the wiki.

aug 30, 2025, 1:22 pm • 3 0 • view
avatar
Gabe Chase @spicoli323.bsky.social

I want to make sure I understand this: "greaterwrong" isn't actually deliberately taking the piss out of LessWrong; it just that it may appear that way? #Poe'sLaw

aug 30, 2025, 3:16 pm • 1 0 • view
avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

it's literally lesswrong, it's a frontend to LW with lower complexity that was started while LW was literally unusably slow

aug 30, 2025, 3:19 pm • 1 0 • view
avatar
Gabe Chase @spicoli323.bsky.social

Huh, I had no idea. Well, now the article context makes more sense, thanks.

aug 30, 2025, 3:23 pm • 1 0 • view
avatar
probably just paul @enpaul.net

> Some problems > Most obviously, prompt injection. "Prompt injection" is quickly becoming to AI what MitM attacks are to crypto: a problem that advocates obsess over because it is technical and therefore engineerable. While ignoring all the actual problems the tech causes for humans in reality

aug 30, 2025, 1:32 pm • 3 0 • view
avatar
Arvid's Computers Are Broken @exys.org

I'm not reading all of this bullshit, is there going to be a pivot show where you pick the funny bits?

aug 30, 2025, 6:14 pm • 2 0 • view
avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

fuck no, the last vid about these bozos was 19 minutes

aug 30, 2025, 6:54 pm • 1 0 • view
avatar
lux @luminousflux.neocities.org

feeding Wikipedia through the "make it worse" machine to create a new, smaller, less accurate Wikipedia

aug 30, 2025, 1:52 pm • 8 1 • view
avatar
AMDO @amdo.bsky.social

and yes it will be costly

aug 30, 2025, 2:00 pm • 5 0 • view
avatar
Kevin Shay @kshay.com

So close to getting it

Another problem is pictures. Many articles benefit from having pictures along the text. [highlighted]The problem is, generated text is text, but a generated photo is by definition a fake photo.[/highlighted] We wouldn’t want the encyclopedia to use fake photos.
aug 30, 2025, 2:32 pm • 2 0 • view
avatar
m0dality @m0dality.bsky.social

This is the stupidest thing I've not finished reading in quite some time. Fucking deranged

sep 1, 2025, 2:04 am • 1 0 • view
avatar
iplayguitarbad.bsky.social @iplayguitarbad.bsky.social

I am now incapable of distinguishing parody anymore. You could tell me this was a good faith attempt and I'd believe you. Likewise, if I got to the end of the page and say "just kidding! This is a dumb idea!" I would nod and say "got me!"

aug 30, 2025, 3:04 pm • 2 0 • view
avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

the rationalists are reliably earnest and the worse the take, the more sincere

aug 30, 2025, 3:17 pm • 0 0 • view
avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

cc @molly.wiki a pain shared

aug 30, 2025, 1:13 pm • 16 0 • view
avatar
Molly White @molly.wiki

beginning to think they may have a point

Screenshot of the linked article, with a sentence highlighted:
aug 30, 2025, 1:51 pm • 34 2 • view
avatar
your #3 source for absurdist true crime 🔨 @davidgerard.co.uk

i mean, the next sentence is the real payoff

aug 30, 2025, 2:00 pm • 18 1 • view
avatar
Glenda Jackson’s Blairite Nation 1997 @joebaldwin.me.uk

Extremely funny given that Wikipedia is functionally a bunch of dispute resolution mechanisms that has somehow made an encyclopedia

aug 30, 2025, 2:06 pm • 21 4 • view
avatar
nick @caveatruptur.bsky.social

A great example of what Fred Brooks was arguing in No Silver Bullet in 1986. The complexity is the essence of the system!

aug 30, 2025, 2:54 pm • 3 0 • view
avatar
Richard K @rkemb.bsky.social

Maybe you update the AI Wiki by just updating the real Wikipedia and waiting for the AI to steal it.

aug 30, 2025, 2:15 pm • 7 0 • view