"Anatol"

Literally. And yet you wouldn't know it from the way the competitive front end of industry for juniors valorises technical skills over social, code golf proficiency over stability and ease of maintenance etc.

not just in this case

if these people had to provide three citations each time they referred to the behaviors of the left, journalism would be over. There would be no more journalism

Escalator has hold me text and hand symbol on rubber bannister strip. I have held it

Providing emotional support to escalator

Letterboxd screenshot Star Trek Into Darkness (2013) Watched by Branson Reese 28 Apr 2022 Abrams isn't a man so much as he's an avatar of cultural entropy. If you make something popular enough eventually you will die or sell it off and and a person in thick rimmed glasses whose main artistic vision is that he loves to have meetings will take it and sandblast it until it's nothing. I hate to see this happen here but maybe it's good that this happened. Maybe every franchise should collapse into a version of itself that makes shareholders nod and shake each others' hands. We should learn to speak in a language that rich people who don't dream can't comprehend.

i often think about this review

Dumping on retail is the ultimate goal, an actual business/product is more of a side effect with an undesirable cost center attached to it

weaponized science fiction. Pioneered by Elon.

And it can't be fixed with a more refined prompt either; "be careful not to fabricate quotes" has no meaning to an LLM. people need to realize it's an inherently dead end "technology"

Journalistic standards? Hi folks, Since Ars is apparently posting partially or fully AI generated articles now, I have to ask - is this going to be a continued policy going forward? That is, will Ars be officially publishing AI generated content from now on? If so, will it be marked? This is obviously pretty concerning.

“Journalistic standards?”

https://arstechnica.com/civis/threads/journalistic-standards.1511650/#post-44249741

So, Ars Technica, a site whose technical commentary is quite popular here, was caught fabricating quotes in a story. What's worse is they don't seem to be able to explain how it happened

Members of the Swarovski family were early, active and enthusiastic champions of Nazism, and at least six of its members maintained membership in the illegal party prior to Austria's annexation to Nazi Germany on 12 March 1938.!23] Three weeks earlier, 500 marchers in the Tyrolean town of Wattens held a torchlight procession that ended with chants of "Sieg Heil" and "Heil Hitler". The majority of the participants, police determined, were Swarovski plant employees, among them Swarovski family heirs Alfred, Wilhelm and Friedrich. 124]

In its report to the state police on 14 February 1947, the Innsbruck district administrator called company head Alfred Swarovski "an enthusiastic member of the NSDAP." [25] Alfred Swarovski praised Hitler at business gatherings and took actions as a regional business leader to ensure that "Tyrolean industry could be integrated as smoothly as possible into the enormous gears of the economy of Greater Germany and into the National Socialist economic order." He sent "grateful loyalty greetings" to Adolf Hitler on his 49th birthday and arranged a donation of 100,000 shillings for Hitler to establish a holiday home in Tyrol.!26]

In 1994, historian Horst Schreiber wrote about Swarovski's past but was not granted access to company archives. (27] The contemporary Swarovski company commissioned historian Dieter Stiefel as "a step towards dealing with our history in a serious and very pro-active manner," board spokesman Markus Langes-Swarovski said in 2018; however, the study was not published because, Langes-Swarovski said, "Swarovski is a company that generally tries to keep the owners' personal stories largely out of the public eye because it does nothing for the business." (28]

Everyone’s talking about Swarovski right now so let’s take a second and dive into the fact that they were NAZI Nazis. Like full-on Seig Heil, prosecuted after the war level Nazis and the modern company has done everything they can to hide all their archives and prevent anyone accessing them.

I just hate informants so much

They prey on your trust, your kindness, your trauma and your neurodivergence to turn you into a weapon against yourself and your comrades for the state while making you feel shame for not allowing them to drive you literally insane

"I say! Are you Georgina?" The child in the opposite bed sat up and looked across at Anne. She had very short curly hair, almost as short as a boy's. Her face was burnt a dark-brown with the sun, and “Famous Five 01 - Five On A Treasure Island” By Enid Blyton 7 her very blue eyes looked as bright as forget-me-nots in her face. But her mouth was rather sulky, and she had a frown like her father's. "No," she said. "I'm not Georgina." "Oh!" said Anne, in surprise. "Then who are you?" "I'm George," said the girl. "I shall only answer if you call me George. I hate being a girl. I won't be. I don't like doing the things that girls do. I like doing the things that boys do. I can climb better than any boy, and swim faster too. I can sail a boat as well as any fisher-boy on this coast. You're to call me George. Then I'll speak to you. But I shan't if you don't." "Oh!" said Anne, thinking that her new cousin was most extraordinary. "All right! I don't care what I call you. George is a nice name, I think. I don't much like Georgina. Anyway, you look like a boy."

To give you an idea of how deranged the UK has become about gender stuff, here's Enid Blyton - the absolute avatar of classic English conservatism and a huge fan of racist "gollywogs" - casually describing a gender non-conforming kid in the 1940s and having their friends be totally cool with it.

Israeli Soldiers Accused of Using Polymarket to Bet on Strikes The country’s internal security agency indicted a reservist and a civilian for security offenses, bribery and obstruction of justice in connection with the bets.

Israel has arrested several people for allegedly using classified information to place bets on Israeli military operations on Polymarket

www.wsj.com/world/middle...

The only benchmark I give a damn about is "OpenAI fires all their HR/lawyers/<insert white collar work>" because they're convinced their AI is good enough to replace them." (Although given the chronic dissemblers and liars they are, even if they did so the second part would still be in question.)

I assumed this was about podcasts.

i love when bsky hides replies on my own post with no way to see them without an external tool, and i only notice it cause i look at my own posts way too often and notice the number of replies doesn't add up

I'd be very skeptical of such claims of discovery. Even AlphaFold has been massively overhyped. The core issue is always that these models have no real understanding.

I think I did at some point, but not entirely sure anymore. Thanks for the reminder, I was aware of the problem but didn't make the connection to previous vaccines, just other built up immunities 🤦🏻‍♂️

Persona Identities Inc (Used by Reddit) The popular forum Reddit uses the San Francisco-based firm Persona for age checks. In 2025, Persona raised $200m in a round led by the Founders Fund, the venture capital group run by data magnate Peter Thiel. Thiel's company, Palantir, has been widely criticised for its ties to the Trump administration and its US surveillance, security, and immigration enforcement contracts.

It’s also funded by real life cartoon villain Peter Thiel

I came pretty close to getting offed by measles at age 18 (had all other vaccines but not that one for reasons unknown - I'm too old for the autism panic). It's not pleasant!

Getting my HP tattoo covered up with Sandman to own the libs

Your profile says "humble". Maybe take the friendly criticism I've seen in the replies to heart.

LLMs are a dead end. No matter how much manipulative lipstick is put on the pig (you're being taken for a ride by a model *optimized to do exactly that*!), they won't ever have the capabilty of learning or reasoning, only a simulacra thereof.

interactions instead of the simulacra an LLM will gaslightingly extrude ("you're right! I made a mistake! *learns nothing*) and I'll be interested. 2/2

Why do tree-based models still outperform deep learning on tabular data? While deep learning has enabled tremendous progress on text and image datasets, its superiority on tabular data is not clear. We contribute extensive benchmarks of standard and novel deep learning met...

I'm not particularly interested in a quiz show, I guess I could point to arxiv.org/abs/2207.08815

But really, this is all incredibly shallow. All these approaches are contrary to how we know actual learning works. Show me a model with an ontology and the capacity to persist feedback from 1/

Counterpoint: you're in a cult

I've been implementing machine learning algorithms since 2005.

There's *nothing* being built with LLMs that's not a giant scam or house of cards. The technology is a dead end and incredibly offensive to anyone vaguely familiar with actual intelligence (which is a hugely complex topic, IQ sucks and is racist, etc, but word extruders sure ain't it, chief)

Anthropic sells LLMs. These are front and center of the current hype bubble and very distinct from other machine learning technologies (which nevertheless all share a common problem, namely that they all require human verification as "correctness" cannot be trained). Don't muddy the waters.

moooo (agreeing)