hendrik

Well, nothing lasts forever. I’d say distributing them on something that lasts 10+ years is better than doing noting. Otherwise they just get lost, buried in the attic or the next harddrive crash takes them.

I think as a country they don’t need “in” on AI. They already do lots of AI, robotics… Published language models that are on par with what’s publicly available elsewhere, use quite some electricity for AI compared to other countries… So I think it’s just the next AI thing, this time from China. And probably part of some strategy to end up in a total surveillance, techno dystopia. Now they need to connect it to WeChat and all of their camera surveillance and they’re almost there.

Yeah, it’s so much novel. And Jarvis - of course - is the only scifi to compare things to.

The article is kind of all over the place mixing high-school graduates and fourth-graders? I can see how you’re sluggish at typing in fourth grade… The numbers for a 17 year old would be interesting… But yeah, 13 words per minute isn’t impressive. And most young people I know use phones and tablets, not computers. So naturally a good amount of them isn’t good around these things.

Yeah we do. But the AI communities here didn’t really take off. Most people here just downvote everything about AI because they’re tired of it. But there are some people who discuss that form of AI that people can run themselves, modify and participate.

Lots of people seem to like them. I’ve read several AI summaries also here on Lemmy. And they kind of always miss the nuance. Oftentimes they entirely miss the point. So much that it sometimes becomes wrong.

I’d blame capitalism. And corporations prefering short term growth and attracting investors. And the whole modern business model of exploiting users private data to sell advertisements. That’s how the whole internet works these days and thak makes being evil baked into every successful company.

And btw: Zuck did one good thing. He personally gave us competetive AI models to tinker around with. If it weren’t for people like him, we would have AI dominate us without the average person having access to more than the online services like ChatGPT. Yeah but that doesn’t take away from the things you lined out.

What’s your bubble of interests? I mean I’ve seen Github projects where that description fits very VERY well. Usually when I’m attending to some very niche hobbies. Or try to get some exotic electronics from 20 years ago running again… With the everyday tools it’s most of the times some active community and I copy and paste the 3 commands and I have it installed successfully.

That’s alright. I learned something anyways. Maybe I’m a bit overly cautios with these companies. I mean they even combine two of the buzzwords that attract investors… “AI” and “Quantum”… But I don’t know much about the topic. Usually we have some classical signal processing available to handle sensor data. But AI can do things like imaging and finding patterns well. I can see how there might be some synergy between the two fields.

you’re dismissing everything […]

Not at all. On the contrary, I’d love to learn some more about it. That’s why I’m asking all these questions. The thing is, we’re talking about something here (quantum computers) and you’re saying they have one. And then you go on talking about an entirely different subject, saying MCG is useful and they bought some SQUID sensors… Of course I’m dismissing that. Since it has nothing to do with the conversation we’re having?! I don’t even disagree. Quantum effects certainly exist. And I bet measuring small magnetic fields is super useful in many applications. But what’s that got to do with the question I asked?

That article talks about a bedside magnetometer device, used to measure the heartbeat of a person. I can’t find any reference to quantum computers 🙁

That certainly counts as hype. But I wonder if there’s any independent information out there about these computers. All I can find is self-advertising and news about investors. I mean we occasionally do get these claims that someone proved quantum supremacy. But as far as I know the validity often isn’t clear or the results aren’t reproduced yet. And sadly I can’t skim the papers since lots of them aren’t open access.

And for research it doesn’t matter if you need days to cool down the computer just for one calculation. Or if most results are wrong due to noise and you have to re-do every computation on a traditional computer to check which results are correct. But I’d expect it takes them years or decades from a protopype like that to something actually useful. And as of now we haven’t even solved superconductivity or the temperatures or decoherence. So I’m always a bit careful with these claims frome the quantum startups.

And does the company you mentioned actually own a quantum computer prototype? Because it seems their focus is writing algorithms/software.

Btw: What a quantum computer can reliably do these days, is tell you 21 is 3 x 7. And it takes hours and quite some traditional computing to do that.

https://en.wikipedia.org/wiki/Integer_factorization_records#Records_for_efforts_by_quantum_computers

We’ve progressed a bit further than that. But for anyone interested in actual applications for quantum computers… They’ll have to wait. It’s research at this point. We’re making progress one step at a time. But so far no one has even demostrated we’re able to scale those computers to a useful size.

So I’d say we’re somewhere close to the origin of the axes. Or on a different graph for research that’s still science fiction. Together with nuclear fusion power plants, thorium cars, space ships and hypothetical battery chemistry that’ll make our electric cars go 5000 miles and not degrade over time.

[Edit: The Wikipedia Article: Quantum comuting also has some good references.]

For info: That’s (just) a SearXNG instance. That’s a metasearch engine, getting results from Google etc and proxying and aggregating them for you.

I think the main issue isn’t directly that, but that they payed $600 billion for things like NVidia hardware, electricity, experts, … And now they need to generate an absurd pile of money. That’s why they might be doomed, because their product now has to fulfill ludicrous expectations. Or the bubble is going to burst.

I agree. Especially music is such a nice thing. And it has quite some power to address emotions pretty directly. And it’s very diverse, individual, multi-faceted. And in theory everyone can participate and learn to play an instrument. Or sing in the car or the shower…

The visuals are the best thing I’ve seen today 😆

I’m pretty sure someone wrote the lyrics here or at least put some good amount of effort in. With my experiments it was far off from that kind of rhymes with ABAB schemes and rhyming a word in the middle of the line.
[Edit: In the video comments they say the text is AI generated, too!]

Sure, back in the day and without AI it was different: https://m.youtube.com/watch?v=9sJUDx7iEJw (RMS singing the Free Software Song).

(I myself don’t have any issues with AI singing and doing music. No one can replace my favorite artists and bands anyways. And I wouldn’t notice if they added yet more soulless and dull pop to the radio program. Because that’s already pretty much there is to it.)

I’m still unsure. That’s certainly a possibility and something that happens in the actual world… Buy a company just for the userbase and throw out everything it consists of. Except for a really tiny portion of the software assets and a few hundred employees. And the database with the user accounts. It’d be super hard to keep the users, though. As they’re then on a platform that’s not anymore what they originally signed up for. If it doesn’t go smoothly, they’ll go someplace else and everything was in vain. Maybe prohibit other private companies from offering competing online services. Or it has to be perfect and stay like that indefinitely.

And I mean the network effect is there. But it can be overcome. Or we’d still use MySpace, ICQ, Facebook, Friendster… I’ve changed instant messenger services like 4 times in my life. Similar for social media platforms and pretty much everything. Just my email is still with the same company.

I’m not entirely sure if that still holds true because companies like Meta and Google are so big these days. But as one example I’d like to mention TikTok which was able to attract like all the young people and get them away from Google and Meta’s grip. And they were able to do that by competing and offering a better(?) service. And it’s pretty much ran by a government. So I’d say it can be done that way. You just need a good product and a lot of money.

But eventually, yeah we should all end up on FOSS services that aren’t paid for in private data.

Do we hate AI now just for the sake of it? I’d say the guitars rock. The singing is shit. But the rhymes are good. But yeah, sounds like Suno AI.

I’d say that’s overly expensive and complex. Since almost everything with these companies is about the ad selling, harvesting and using the data and tieing the users attention. The state would adopt something that is mostly concerned with that. And they’d struggle with their role influencing political views with the algorithms that now belog to them. And it’d be pretty much an Orwellian dystopia once the state starts getting into the advertisement business. What we consider a “product”, the social media platform or mail service is just a means to have users. It’s a tiny fraction of what these companies do. And it’s an expense to them, not what they make money with.

I think it’s far easier and quicker to start fresh. Have something that has good features baked in from the start. And not adapt a business, settle >90% of what it’s about and change the product 180 degrees so it’s about something entirely different. I mean everything Google programms is with the idea in mind to sell ads. They’d need to change pretty much everything about that program code. And we already have some good alternatives to some things. And the EU for example already funds some Free Software. I think if we were to educate people, regulate online services in a good way and offer proper alternatives, the rest follows automatically. IMO nationalising an ad selling business comes with severe issues, as I lined out earlier. And if we did it over, we could also learn from the past and address issues like filter bubbles, unhealthy behaviour, being overly addictive and whatever is baked in to the current generation of social media and almost impossible to get rid of.