I’ve read that article. They used something they called an “MRI for AIs”, and checked e.g. how an AI handled math questions, and then asked the AI how it came to that answer, and the pathways actually differed. While the AI talked about using a textbook answer, it actually did a different approach. That’s what I remember of that article.

But yes, it exists, and it is science, not TicTok

Only if the also take the full legal responsibility for the AIs actions.

It does. Shitty admins do shitty things, regardless of platform.

Just get over Reddit. Forget it, and start anew on greener pastures.

If a business cannot survive without breaking the law, then it is not a business but a criminal organisation.

Indeed. Simply that. If a business is not sustainable without breaking the law, it is not a business, it’s a criminal organisation.

At the moment, what they call AI is still just an LLM, a Large Language Model, or, in other words, a parrot with a large dictionary.

It is anything but smart.

He simply does not like it because it does not fill the billionaires pockets. Whether he calls it “racism”, “terrorism”, or “gobbledigook” does not make a difference.

He still has absolutely no clue how tariffs work. Amazing for someone who throws them left and right without thinking. Although the latter might explain it.

If you draw a sketchup that wrong, you better hire a professional.

I would not want them to monitor my calls, either. I meant exactly enforcing SHAKEN/STIR - if they cannot guarantee that the call/text is from the number shown on the display, don’t route it.

This way you can block the spammers or even sue them if you are on a do-not-call list.

It should be the other way round. Telco providers should be fined for any spam call and -text they don’t prevent from happening.

I run a computer on Win7 at work, because it needs some important legacy software. It can’t be containered because it has a nasty licence manager.

And my oscilloscope runs on Win98.

Lure them in, pull the rug, harvest profits.

You’re right, cost of content creation comes on top of the running costs, too.

With Taler I wait for it to work the kinks out. And, of course, for it to be available somewhere reasonable.

But even running an indexer on a YT-like scale would need serious money, even if you spread the hosting and streaming load around. And for most users, this would not be attractive, as you probably would have to torrent the data first and view it later.

Then there is the issue with responsibility. If someone throws e.g. CSAM into the system, who could be held responsible? Who would have to deal with DMCA notices? Who would deal with issues like “Dictator X demands all videos showing him in a bad light to be removed immediately!”

And: Opening a payment system is a serious can of worms, especially if you need it to work internationally.

Honestly, I’m not against a YT alternative, but I don’t want it to die after three weeks because the person behind is was too optimistic to consider to potential problems.

The core problem is that hosting and streaming videos costs money, and that money must come from somewhere. Unless there is someone with really deep pockets just paying for everything, such a platform must use subscriptions or ads to make some money. Netflix & others use subscriptions, YouTube uses ads, and both even offer combo models.

How would a free variant of YouTube work on the long run? Setting up a small model on a server in your home office, maybe with donations to cover initial hardware costs is not the issue at first, but once you need a computer center and employees you’ll need some serious, regular money coming in.

Simply abandon the US and move the forum to civilization.

They show me an ad, and they are out. It is that simple. Luckily, my life does not depend on streaming.