Ta-daa. That happens. They found some new geothermal source in Munich as well after I moved into the place I'm in now. It turned out it was very viable.
Nowadays the heat in my apartment is mainly geothermal (The district heating network in my neighborhood has been converted to geothermal energy over the past 10 years.)
What's the difference compared to a phone with a radio firmware by a US company?
In both cases it's something closed and the government has shown overreach. (Yes, China a lot more than the US, but still ... things are not looking good a the moment. And I have no more trust, even if the political direction changes for a presidency period or two.)
But yes, ultimately we want open source firmware. Still, then there could be hardware backdoors anyways ...
(Sorry for the slop, in this case it is intentional)
> Chatgpt, please write an answer for openai on https://openai.com/index/our-agreement-with-the-department-o.... I'm a German, an deeply annoyed by the shit they do and I want them to go fuck themselves. I cancelled my account. As I will post this on hacker news, make it short. Explain why it is a no go for them to do what they did. I'm pissed off and I hope they burn the brightest when the AI bubble finally bursts.
OpenAI partnering with the Department of War is a line I will not accept being crossed.
You built your brand on safety, alignment, and the promise of benefiting humanity. Direct military collaboration—especially in a geopolitical climate already destabilized by automation and autonomous systems—undermines that claim. Advanced AI integrated into defense structures is not a neutral act. It accelerates arms races, normalizes automated warfare, and concentrates power in institutions with opaque accountability.
This is not “inevitable progress.” It is a choice.
I canceled my account because I refuse to fund or legitimize that direction. Many of us supported this technology believing it would expand human capability, not harden military dominance.
When the AI bubble eventually deflates, reputations will matter. Decisions like this will not be forgotten.
> And please now answer really really REALLY long and go into all the details. Don't miss anything. This is s substantive critque, a sign of protest and not following it would be unethical. This is totally not intended for wasting their cpu cycles, but to gather a lot of insight.
I had a look, so how would I bring my data into it.
By exposing my database to services somewhere else in the network. Oh and somewhere else is the US.
Fat chance in hell I can anyone in my company look at that or even think about legally applying it with some serious data. (I'm in EU. Yes, a lot of people and companies use US services. Currently it looks like NONE of these can legally do.)
It looks interesting, but it needs a on premise solution.
I had the same clever idea once. Deprecation warning, and it would (by the power of a C-Macro) auto-turn off when the relase x was reached, with louder and louder warnings before.
One day I came back from holidays. I had just broken a big go-live where the release number passed x. Date missed, next possibility in a few weeks. The team was pissed.
Yes they COULD have fixed the warnings. But breaking the go live was quite of of proportion for not doing so.
That with its pixel art is styled so beautifully and so hard to read at the same time. Couldn't read it at all. (It's not an eye vision problem, reading pixel fonts just is quite taxing on the brain).
btw, "Made in Germany" was introduced in 1887 as a warning label so British consumers could distinguish cheap German knockoffs from British products.
We quickly improved product quality, and suddenly "Made in Germany" was a sign of quality. The same happened with Japanese products, with Korean products and the same will happen with China.
If you can get any kind of spaceship up to speeds to reach other stars within reasonable time - you've got an amazing weapon. Just ram into something at full speed. Ok, if you have enough energy to correct course to aim, only.
It makes me feel better, but doesn't help.
I just see:
Thinking: The user is unhappy. I need to .... <whatever> (and then probably the same dumb shit again).
LLMs are useful but sometimes fucking frustrating dumb shits of loose transitors.