OhNoMoreLemmy

Yeah, and you can really see this in image generation. There's often blocks on using the names of celebrities in the prompts, but if you misspell the names enough it can bypass the censor, and the image generator still understands it.

Other way round. The nickname/insult was saying Germans are warlike barbarians like Atila the Hun and the rest of the the Huns.

You might want to look at Wittgenstein.

In his early work he went hard on this approach, and insisted that "hey philosophy is dumb", just agree on the definitions and then chase through the implications.

In his later work he realised that this is impossible. Words have contextual meaning that is revealed by their usage and you can't nail down full and complete definitions in advance.

What you're talking about absolutely can and will never work. We have tried it and seen it fail.

To be clear they're both shit from a privacy perspective.

Telegram has effectively no security by default. WhatsApp has better security on paper with meaningful end to end encryption. But Facebook still get your meta data and they scan images in the chat to check for illegal data.

That said I basically agree with you for insecure conversations. Telegram might have access to more of my data in a chat but they don't connect it up with all of the data Facebook has tracking me across the internet.

If Musk doesn't pay the loans the banks can take Twitter off him, but then they're stuck with Twitter.

It's always lost money, but now Musk has driven off the advertisers and many of the high prestige users; got stuck in a bunch of pointless lawsuits he's going to lose; and run up a lot of debt by refusing to pay people.

And Musk knows this. The banks are fucked.

Yeah it's super obvious that he's a product manager from the quotes.

the executive said. "The skill in and of itself is like, how do I innovate? How do I go build something that's interesting for my end users to use?"

This is the path for promotion for product managers. Create new interesting products and move on before they fail. And yeah, if you really don't care about failing LLMs can maybe help speed up prototyping here.

However, what Aws actually tries to sell to users is rock solid reliability and high up time. If you start asking "how do I go build something that's even more reliable?" It's incredibly clear that LLMs are not the answer.

As someone working and publishing in the field this is more a cyber jerk about American exceptionalism than actually true.

Chinese universities and companies publish a shit tonne at pure machine learning conferences. They absolutely do a large amount of research into the fundamentals of machine learning as well as the applied stuff. They're probably the closest to the US in terms of having large firms that are prepared to bank roll the training of the very large language models.

Alibaba in particular has been constantly doing cutting edge stuff in terms of multimodal language models that are worth paying attention to.

The actual truth is that China does both kinds of work. Broad foundational and applied work lead by independent research groups in companies and universities, and focused application driven stuff for direct application by the state.

Google still stands out in terms of the amount of research it does, but this is because Google is different to everyone -other US research institutes don't compare to it either.

Can he drop out?

Kamala had to be nominated before the convention to make registration deadlines in Ohio.

Shouldn't RFK be stuck on the ballet? - at least in Ohio?

It's going to be so much worse than pasting ads if it ever actually works.

Rewriting the search results with LLMs let's you seamlessly mix the adverts into the text people are trying to read... That's why I need to sip a refreshing glass of coke zero(TM) when I'm contemplating the collapse of the internet into pure advertising drivel.

Amazon are also dicks about sick leave. I'm sure forcing people to work, and work hard, when they're ill leads to complications.

No. The problem with your current bot isn't that the website authors have a particular axe to grind, it's that they're just in a rush and a bit lazy.

This means that they tend to say news sites which acknowledge and correct their own mistakes have credibility problems, because it's right there - the news sites themselves acknowledged issues. Even though these are the often most credible sites, because they fix errors and care about being right.

Similarly the whole left-right thing is just half-assed and completely useless for anyone that doesn't live in the US. While anyone that does live in the US probably already has an opinion about these US news sources.

Because these are lazy errors, lots of people will make similar mistakes, and aggregating ratings will amplify this, and let you pretend to be objective without fixing anything.

The app phones home to access recipes.

Maybe it works without access to the server, but maybe it just refuses to do anything.