You know how sometimes a story seems so completely crazy, that you wonder what detail has been left out? Like… the payment provider just randomly wanted a list of all passwords? What?

What Intel giveth, Microsoft taketh away.

Yes. It’s almost as if that wasn’t the point I was making at all.

If homeless people were allowed to stay in the spaces that they’re currently being chased away from, there’d be a lot fewer homeless people.

I predict yet another Signal-related hack within the month.

They keep saying “stolen” when their model hasn’t even been directly copied, just used to train another.

The other pathetic thing they’re doing is pushing the narrative that it’s being done for weapons and spying and shit, as though that’s exclusive to deepseek. Sadly at least a few of the useful idiots on my LinkedIn seem to be swallowing and regurgitating it.

me too, it’s a slightly awkwardly constructed sentence.

It’s a UX issue. If you make votes public they should be very clearly so, the way reactions are on Facebook, LinkedIn etc, and everyone would be fine with it. What’s getting people’s panties in a twist is that Lemmy superficially presents itself as having anonymous voting but the logs are tucked away behind some convoluted process that you have to just know about and only seems to work about half the time.

I used to think it was weird the way he pronounced the “s” in “Asia”, and then I randomly heard it in some other old film and realized that people just used to say it like that a century ago and he had no reason to change now.

I read it as criticising reductionist views of the many diverse nations that existed in North America before Europeans showed up and decided that the whole continent was Terra Nullius.

To this day a significant number of US high school American History textbooks only discuss the tribes in terms of their interactions with European invaders, and shy away from anything that might make them look like they were ever legitimate nations. Referring to them as ‘Native Americans’ instead of by name also has this effect.

That’s my whole point, it wasn’t Ballet Vision’s option to exercise, it was Kibbutz Hanita’s, but the opening summary erroneously describes it as the exact opposite. Here’s from later on in the article:

As part of the agreements, the Chinese fund granted the remaining minority shareholders an option to require it to purchase their remaining shares for about $9.5 million, now valued at roughly $11 million, by early December 2025.

https://www.investopedia.com/terms/p/putoption.asp

(edit: I had a little search just now to see whether it was true that China have made it illegal to invest in Israeli companies. Half the results were literally reporting on this story, and the other half of results were about Chinese companies actively aiding illegal Israeli settlements, which makes me suspect that BV’s claim that their hands are tied is bullshit.)

accusing it of refusing to exercise an option

Was this written by an AI or by an illiterate? That’s not what the rest of the article says happened. The Chinese fund didn’t refuse to exercise the option, they refused to honour an option that they’d already signed. The opening sentence makes it look like Kibbutz Hanita are pulling a fast one, when it’s the exact opposite.

I wonder how true it is. Humanoid robots are popular in sci-fi because you can write lots of interesting stories with them, and it’s easy to film since you can just use a regular actor. But if I get to thinking what I’d want out of having one in the house, I’m drawing a blank. I’d rather they just made Roombas and dishwashers better.

The insane thing about TSLA’s valuation compared to either its profit or revenue, is that it’s based on the idea of it growing by a bigger margin than the market could possibly have room for, even if they did turn around their tired-out product lineup and managed to get everyone to stop hating them. People still seem to treat them like a plucky new startup in its fast growth phase, even though they’ve been around two decades already.

They made barely 0.2% of their market cap in profit in 2025. A 12 month US treasury bond pays 3.52%. TSLA could increase profits by over 1500% and still be worse than a treasury bond whilst also being higher risk.

deleted by creator

The hash proves which bytes the answer was grounded in, should I ever want to check it. If the model misreads or misinterprets, you can point to the source and say “the mistake is here, not in my memory of what the source was.”.

Eh. This reads very much like your headline is massively over-promising clickbait. If your fix for an LLM bullshitting is that you have to check all its sources then you haven’t fixed LLM bullshitting

If it does that more than twice, straight in the bin. I have zero chill any more.

That’s… not how any of this works…

How does having a key solve anything? Its not that the source doesn’t exist, it’s that the source says something different to the LLM’s interpretation of it.

I haven’t tried this tool specifically, but I do on occasion ask both Gemini and ChatGPT’s search-connected models to cite sources when claiming stuff and it doesn’t seem to even slightly stop them bullshitting and claiming a source says something that it doesn’t.

Hey, you’re the one underpaying your fictional workers in the first place

srsly though it’s astonishing just how much global geopolitics hinges on whether this is successful. MS Office is the main blocker to the EU migrating off US tech.