Live data from Hacker News

Viewing profile — phoe-krk

phoe-krk

HN member
Joined
Fri, Feb 10, 2017, 1:41 PM UTC
HN karma
9,248
Public activity
1,454 items

About phoe-krk

Lisp hacker. https://phoe.github.io

Author of "The Common Lisp Condition System", published by Apress. https://www.apress.com/us/book/9781484261330

meet.hn/city/50.0619474,19.9368564/Krakow Socials: - github.com/phoe - functional.cafe/@phoe ---

Recent public activity

  1. comment
    Comment #49216261

    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, becau…

  2. comment
    Comment #49215999

    > 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 w…

  3. comment
    Comment #48944379

    There is a secret in the code though, minified enough to be illegible.

  4. comment
    Comment #48643970

    Possible context at https://en.wikipedia.org/w/index.php?title=Wikipedia:Adminis... - might be a better link, too.

  5. comment
    Comment #48459623

    And now imagine that one of the Trump tantrums contains an announcement of sanctions against the European Union.

  6. comment
    Comment #48235539

    Screenshots of code? In 2026?...

  7. comment
    Comment #48209053

    Just substitute asm.js with WASM and you're still on the right track.

  8. comment
    Comment #47940980

    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, hum…

  9. comment
    Comment #47731623

    > 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 "c…

  10. comment
    Comment #47731282

    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 ei…

  11. comment
    Comment #47728493

    > 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.

  12. comment
    Comment #47728223

    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…

  13. comment
    Comment #47577329

    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.

  14. comment
    Comment #47576794

    You can try it now! Prices starting at €13.99 per month, billed yearly.

  15. comment
    Comment #47575732

    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 br…

  16. comment
    Comment #47540880

    > Unlike perl, sbcl doesn't even pretend to care about security. Mind expanding? What particular stuff does Perl have in terms of security here?

  17. comment
    Comment #46995212

    > 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-a…

  18. comment
    Comment #46912229

    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 …

  19. comment
  20. comment
    Comment #46841615

    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 overdelegatio…

  21. comment
    Comment #46820775

    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 st…

  22. comment
    Comment #46816694

    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 ot…

  23. comment
    Comment #46816396

    > "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 to…

  24. comment
    Comment #46645156

    > 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 …

  25. comment
    Comment #46630109

    Huh, https://looooooooooooooooooooooooooooooooooooooooooooooooooo... comes to mind. Thanks!