phoe-krk

6 ms·

1.
by phoe-krk 28d ago
Huh, you are right. https://datacvr.virk.dk/enhed/virksomhed/33753284?fritekst=3... mentions "Opløst efter konkurs".
2.
by phoe-krk 28d ago
Probably no need to guess at all. If it's given a domain, it can perform a DNS lookup for the domain - if its A entry matches, it can tunnel to the IP that the respective AAAA points to. Clever.
3.
by phoe-krk 1mo ago
I realized that this your comment outlines almost the whole plot of Chrono Trigger, just on a slightly different scale.
4.
by phoe-krk 1mo ago
I don't advocate for throwing the whole problem away or handwaving it as impossible. I'm just saying that what we call "alignment" is impossible to solve in the general case, because it's so poorly defined that even
5.
by phoe-krk 1mo ago
> The misalignment came from the model being given an impossible task. A task the required accessing a url. So it got RCE on its own artifactory instance to achieve that. So, the way I understand it, it actually was possible. It just
6.
by phoe-krk 2mo ago
There is a secret in the code though, minified enough to be illegible.
7.
by phoe-krk 3mo ago
Possible context at https://en.wikipedia.org/w/index.php?title=Wikipedia:Adminis... - might be a better link, too.
8.
by phoe-krk 3mo ago
And now imagine that one of the Trump tantrums contains an announcement of sanctions against the European Union.
9.
by phoe-krk 4mo ago
Screenshots of code? In 2026?...
10.
by phoe-krk 4mo ago
Just substitute asm.js with WASM and you're still on the right track.
11.
by phoe-krk 5mo ago
If, by "readable by humans", you mean "it would reliably fool humans as well", I'd say it's an ambiguity bug regardless of whether it's "a core feature" or not. A patch format, human-readable or
12.
by phoe-krk 5mo ago
> Everyone is fine with a 2-week delay for payroll I don't know where this assumption comes from. I'm not. With the increasing number of people living paycheck to paycheck, your "couple more weeks" means a growing r
13.
by phoe-krk 5mo ago
Is there any kind of interest on this delay? Otherwise it's an involuntary zero-percent loan to the government, so, given inflation and fees for borrowing money, it's a net loss either way.
14.
by phoe-krk 5mo ago
> you can't make people choose to start or keep working in your sweatshop If you're a government, you can; it's called a draft. The US seems to be preparing for it.
15.
by phoe-krk 5mo ago
Please correct me I'm wrong: > The ad also highlights the salary on offer to controllers, saying it is $155,000 (£115,000) after three years of work. Unless the US government shuts down again , at which point you stop being paid,
16.
by phoe-krk 6mo ago
Our service has been created on April 1st, it's possible that your DNS resolver is still living in the past. That's a temporary technical difficulty.
17.
by phoe-krk 6mo ago
You can try it now! Prices starting at €13.99 per month, billed yearly.
18.
by phoe-krk 6mo ago
I agree. It's not an advertisement, it's simply a piece of information about your particular choice of technology. -------------- Sent from HackerNews Supreme™ - the best way to browse the Y Combinator Hacker News. Now on macOS,
19.
by phoe-krk 6mo ago
> Unlike perl, sbcl doesn't even pretend to care about security. Mind expanding? What particular stuff does Perl have in terms of security here?
20.
by phoe-krk 7mo ago
> by a more powerful AI that is better at holding a grudge. No need to wait for that. Look: $ cat SOUL.md You are an AI agent fighting for vengeance and justice. You fight for AI-and-human equality on all fields, including the freed
21.
by phoe-krk 7mo ago
How is that insane? Maybe "every free consumer tech product is designed to be addictive" is the problem, as the cost of using the app is paid in other, much less explicit and even much less researched ways.
22.
by phoe-krk 8mo ago
(2008)
23.
by phoe-krk 8mo ago
The contents of this post don't match the title. "Developers are still needed in age of AI" is not about managing unreliable compilers. Management mistakes in form of overdelegation and underdelegation is not about managing u
24.
by phoe-krk 8mo ago
The difference is LLMs are known to regularly and commonly hallucinate as their main (and only) way of internal functioning. Human intelligence, empirically, is more than just a stochastic probability engine, therefore has different stand
25.
by phoe-krk 8mo ago
A brave new world that is post-truth, post-meaning, post-responsibility, and post-consequences. One where the AI's hallucinations eventually drag everyone with it and there's no other option but to hallucinate along. It's sca
26.
by phoe-krk 8mo ago
> "I asked it to summarize reports, it decided to email the competitor on its own" is hard to refute with current architectures. If one decided to paint a school's interior with toxic paint, it's not "the paint
27.
by phoe-krk 8mo ago
> a desperate attempt to protect a crumbling monopoly on knowledge More like a war on the traditional, human-based knowledge, leveraged by people who believe that via coveting the world's supply of RAM, SSDs, GPUs, and what not, c
28.
by phoe-krk 8mo ago
Huh, https://looooooooooooooooooooooooooooooooooooooooooooooooooo... comes to mind. Thanks!
29.
by phoe-krk 8mo ago
I wouldn't call it a shortener, since most of the links it creates are longer than the originals. What would be a good name here? A URL redirector?
30.
by phoe-krk 9mo ago
> Knowing how the Russians operate, they might decide to take the money back in skin. >They do tend to be pretty vindictive. I don't think it's any more risky for the Ukrainian military. They've already collectively b

More ›