If they can refinance their own customers to buy their own products without getting caught in the middle that would be even more scandalous than the circular financing itself but then, the financial system has been completely broken for a good while.

That extortion has an end date, hopefully 2030, when the EU’s Iris2 gets operational. Until then however Elon can play his nasty game, at least for the higher quality level of the service.

Yes, it is. Their costs aren’t going down, only what they can charge and they struggle to be profitable already now. Their only hope is outlawing open models from frontier applications and coding.

You do, however Nvidia got greedy and started to circular invest in the billions if not trillions into those buying the shovels. Now, when the bubble bursts, Nvidia will be hit massively as well.

Is this finally the crack, or will they manage to paper over it and keep the bubble growing even further?

They are really making an increasingly strong case for local open weight models, run either on local infrastructure or rented data centre infrastructure. How long until they are going to try making those local models illegal?

All the tech bros were constantly looking down on ESA, Arianespace and Ariane 6. Demanding that this perfectly good but slightly more costly platform should be thrown away and that Europe should make itself completely dependent on SpaceX+Russia instead. ESA and EU did not. Surely SpaceX would have gladly launched Iris2, the sovereign strategic alternative to Starlink. Right? Right?!

Noyb is doing such such an important work.

That’s fair. It still shows how far you can get even with a foldable phone when you are taking care of it. It also shows how long these things are already on the market. Damn, time flies.

5 years for a fold3? I am impressed.

I do believe there is a target audience for folding phones but if you had already "multiple of them, you don’t seem to require a device that can truly last (+5 years).

I have no idea if they did it in NK at any point. I am pretty sure it is nothing you can see in current North Korea. It is a bit older already but in 2008 two Austrians were travelling by rail, via Russia to North Korea. They entered the country via north by train in a way, that the North Korean regime officially allowed at that time but apparently had no process for to handle in any way, because they probably did not expect anyone crazy enough to take the full Transsib to North Korea. What they managed was travelling by train in North Korea, all the way from the very north, to the capital Pyongyang, where they were then greeted by their “guides” as usual for Western tourists.

They took a lot of pictures and videos of rural North Korea and its people and they had uncensored contact with regular North Koreans on the train, who were understandably very intrigued by those very strange people from abroad. I think to this day, the material they took is among the best, uncensored and non-biased evidence of life in North Korea. Limited as it may be from simply travelling by train across the country.

https://youtu.be/b1mzXFAzUjQ?t=694

I am not saying she is not a serial liar who is earning her money from selling out to the populist right media, but you’d be surprised what humans can do in terms of pushing rail carriages, as long as the track is flat. Rail has very low friction. For a single person it is very hard, but if you have a dozen of people at it, things get doable.

https://www.youtube.com/watch?v=9qGJFDyrGEg

I could imagine that shunting work of individual carriages could be done by human force, if you have enough people available for it. With maybe two dozen people for one carriage it could be even not totally exhausting work. Still a rather crazy thing to do but perfectly possible.

In the US you can hide that opt-in in unreadable terms, can’t you? In the EU that will be harder though as there the opt-in has to be easily readable and obvious to the regular user, as well as feature an option to decline without making things non-functional that do not need that for their very functioning.

In any case, people will be surprised how inshittified AI will be able to become once they have done the same they have done to regular search.

I had not heard of that. I think to stay correct, that needs some qualification. They have given up to compete with Claude on coding, so for agentic stuff the “Grok Bot” is using also Claude Opus 5.5. Grok still is an option too.

But yeah, it does sound like they are slowly giving up to compete at the very frontier.

The outlook of Tesla isn’t great, so it is a perfect fit for SpaceX+xAI. All the brand new models are burning trash heaps, there are two cash cow models increasingly aging, even with the face lifts. It makes sense to consolidate everything into one big grift conglomerate.

This is an inherent flaw of digital voting machines. One has to trust their integrity which only can be confirmed by a few specialists. By design, that is a lot simpler to undermine than a system that relies on a huge number of people where almost anyone who wants to is capable of taking part in securing the integrity of the vote.

No self respecting democracy should use them.

It is the company’s own content. They are running the LLM and the LLM is just a tool with no responsibility on its own. If the tool is unreliable that is Google’s problem.

I am a bit confused. Which institutional EU website features auctioned ads to begin with?

The point is that it could be worse, considerably worse. Yes, I know, the standards are low nowadays. I try to stay with GOG where possible but with Linux, you sometimes have better compatibility with Steam versions and you have the steam DB reducing the risks even further, even if that is not from Valve itself.