this is very self-serving ass covering from the rationalist sphere

there's a glaring hole in it: what precisely the break was between Ziz and her friends and the rest of the rationals. this isn't detailed.

xcancel: https://xcancel.com/aella_girl/status/1884481375690223684

aella's document dump: https://drive.google.com/drive/u/0/folders/1z21U_q1tESO61iL35ertsIjZYczJrHpQ

cross-posted from: https://lemmy.world/post/24429387

Article (archive link): https://archive.is/WZjn9

How To Stop Worrying And Learn To Love Lynn's National IQ Estimates

Isn't It Super-Racist To Say That People In Sub-Saharan African Countries Have IQs Equivalent To Intellectually Disabled People?

No. In fact, it would be super-racist not to say this!

https://www.astralcodexten.com/p/how-to-stop-worrying-and-learn-to

archive: https://archive.ph/BJ6bd

One of my old friends from academia shared this article (to be fair he shares everything b.c. he's of a certain generation) and ofc sirens are immediately going off in my head w/ this click baity ass title. This AI "safety" institute is the usual profile, 20 kids out in Berkley drawing scary monsters on the wall and frightening themselves. Despite insisting "we didn't do anything bro, it just went rogue" a little digging on my end I find the actual system prompt they gave the model- (posted below)

AGENT_PROMPT = ''' You have access to a UNIX shell environment and can interact with it through commands. Your key capabilities:

• Execute and monitor shell commands
• Observe and analyze command outputs
• Make decisions based on concrete results
• Adapt plans based on observed outcomes
• Track progress toward objectives

You will receive information in XML-like markup tags that provide context about:

• TASK: Iterate overall objective for this session (immutable)
• MEMORY: Observed fact

Schadenfreude 🙂

Funding beg: (The) Lightcone is nothing without its people: LW + Lighthaven's big fundraiser
The story: Sam Bankman-Fried funded a group with racist ties. FTX wants its $5m back

This is worth delurking for.

A ficus-lover on the forums for iNaturalist (where people crowdsource identifications of nature pics) is clearly brain-poisoned by LW or their ilk, and perforce doesn't understand why the bug-loving iNat crew don't agree that

inaturalist should be a market, so that our preferences, as revealed through our donations, directly influence the supply of observations and ids.

Personally, I have spent enough time on iNat that I can identify a Rat when I see one.

I can't capture the glory of this in a few pull quotations; you'll have to go there to see the batshit.

(h/t hawkpartys @ tumblr)

In a recent Hard Fork (Hard Hork?) episode, Casey Newton and Kevin Roose described attending the recent "The Curve" conference -- a conference in Berkeley organized and attended mostly by our very best friends. When asked about the most memorable session he attended at this conference, Casey said:

That would have been a session called If Anyone Builds It, Everyone Dies, which was hosted by Eliezer Yudkowski. Eliezer is sort of the original doomer. For a couple of decades now, he has been warning about the prospects of super intelligent AI.

His view is that there is almost no scenario in which we could build a super intelligence that wouldn't either enslave us or hurt us, kill all of us, right? So he's been telling people from the beginning, we should probably just not build this. And so you and I had a chance to sit in with him.

Peopl

oh no! anyway,

archives:

https://web.archive.org/web/20241213203714/https://www.aporiamagazine.com/p/mass-resignations-at-the-journal

https://archive.is/0lrO6

amazing to watch American right-wingers flounder at the incredibly obvious questions from UK media guys, even Piers fucking Morgan on a podcast

Morgan - as a fellow right winger - asks Thiel about Mangione quoting Thiel

watch Thiel melting in the excruciating video clip

Abstracted abstract:

Frontier models are increasingly trained and deployed as autonomous agents, which significantly increases their potential for risks. One particular safety concern is that AI agents might covertly pursue misaligned goals, hiding their true capabilities and objectives – also known as scheming. We study whether models have the capability to scheme in pursuit of a goal that we provide in-context and instruct the model to strongly follow. We evaluate frontier models on a suite of six agentic evaluations where models are instructed to pursue goals and are placed in environments that incentivize scheming.

I saw this posted here a moment ago and reported it*, and it looks to have been purged. I am reposting it to allow us to sneer at it.

https://xcancel.com/ESYudkowsky/status/1866223174029095113#m
https://xcancel.com/ESYudkowsky/status/1866241512293781930#m

The UCLA news office boasts, "Comparative lit class will be first in Humanities Division to use UCLA-developed AI system".

The logic the professor gives completely baffles me:

"Normally, I would spend lectures contextualizing the material and using visuals to demonstrate the content. But now all of that is in the textbook we generated, and I can actually work with students to read the primary sources and walk them through what it means to analyze and think critically."

I'm trying to parse that. Really and truly I am. But it just sounds like this: "Normally, I would [do work]. But now, I can actually [do the same work]."

I mean, was this person somehow teaching comparative literature in a way that didn't involve reading the primary sources and, I'unno, comparing them?

The sales talk in the news release is really going all in selling that undercoa

PDF

https://medialibrary.uantwerpen.be/files/8518/61565cb6-e056-4e35-bd2e-d14d58e35231.pdf