catigula

15 ms·

1. 361.▲
   by catigula 1y ago

   If you think that's crazy, it's likely a drop in the bucket comared to the noumenonal world. There's no reason to think that our senses encompass the vast majority of understanding everything in reality and current evidence t
2. 362.▲
   by catigula 1y ago

   You get a new device every year and teams of professionals are constantly churning updates for it. With cars, you don't get to get a new device, it has to be consistent and keep working and you had better make it all work with a skelet
3. 363.▲
   by catigula 1y ago

   This talk of arms race with China is very, very dangerous. We need to start having a more important conversation if people really believe this because, otherwise, why shouldn't China just kinetically strike US infrastructure and vice v
4. 364.▲
   by catigula 1y ago

   I made two basic arguments: 1. The exponential curve of AI progress 2. The lack of any mechanism that will immediately wall off AI progress at this specific exponential part of the curve >I don't see how AI is going to replace, say,
5. 365.▲
   by catigula 1y ago

   Unfortunately this is an argument from incredulity.
6. 366.▲
   by catigula 1y ago

   Whatever the power/expense GPT-4 -> GPT-5 is such a large improvement that I can scarcely say that it is very different from GPT-3 -> GPT-4. If this happens only one more time, you're going to see really large changes.
7. 367.▲
   by catigula 1y ago

   >Some estimates suggest that half of all white-collar jobs will disappear as artificial intelligence advances I find this so odd. Is there an imagining that there is some imminent trivial threshold at which AI will stop improving? If AI
8. 368.▲
   by catigula 1y ago

   Not really. After using GPT-5 extensively I'd easily trust it to do my taxes. I find the output to be very reliable, also my taxes are trivial.
9. 369.▲
   by catigula 1y ago

   I feel like you're attesting to interior knowledge about a LLM's state that seems impossible to have.
10. 370.▲
    by catigula 1y ago

    I think what you will find is that many people fundamentally don't care about their job for various reasons, chiefly of which is most likely that they don't feel fairly compensated, and thus outsourcing their labor to AI isn&#x2
11. 371.▲
    by catigula 1y ago

    Tough to gain any predictive information here due to the anthropic principle requiring a series of comical happenstance for observation to even occur.
12. 372.▲
    by catigula 1y ago

    Kind of concerning when you see governments hardening themselves to massive AI cyber attacks.
13. 373.▲
    by catigula 1y ago

    Thanks
14. 374.▲
    by catigula 1y ago

    I found myself holding some critical feedback back recently of a poor management decision due to caution about my career. My co-worker, who is more matured in his career, with less need for caution, instead took up my complaints (a viewpoin
15. 375.▲
    by catigula 1y ago

    What is "flamebait"? Unpopular beliefs I genuinely hold? I looked it up and the definition seems to include being disingenuous, which I am not. Claiming I am is annoying and feels like it is a personal attack.
16. 376.▲
    by catigula 1y ago

    [flagged]
17. 377.▲
    by catigula 1y ago

    I'm still absolutely right constantly, I'm a genius. I also make various excellent points.
18. 378.▲
    by catigula 1y ago

    I happened to be in the middle of a task in a production codebase that the various models struggled on so I can give a quick vibe benchmark: opus 4.1: made weird choices, eventually got to a meh solution i just rolled back. codex: took a di
19. 379.▲
    by catigula 1y ago

    They aren't on the back-foot, it's just that there's no way to minimize civilian casualties when Russia sends weapons of war into neutral civilian zones by definition. The answer becomes deny and don't engage. It's
20. 380.▲
    by catigula 1y ago

    [flagged]
21. 381.▲
    by catigula 1y ago

    A simple solution is to hold all chess matches in a SCIF.
22. 382.▲
    by catigula 1y ago

    [flagged]
23. 383.▲
    by catigula 1y ago

    I'm glad you realize how silly that word has become. In reality, groups of people via culture or whatever other mechanism do generate certain things that are undeserving or deserving of censure. For example, due to cultural reasons, 19
24. 384.▲
    by catigula 1y ago

    I think we're stuck and have to agree to disagree but the message sent is at least indistinguishable from the message of a self-interested sociopathic community with no moral concerns beyond their own. When I do things I at least try t
25. 385.▲
    by catigula 1y ago

    No, that isn't my general reaction to Hitler saying he killed Jews to make Europeans safer. My reaction is "what are you talking about, psycho murderer?" That's a good look, try that.
26. 386.▲
    by catigula 1y ago

    Desperation for new data harvesting methodology is a massive bear signal FYI
27. 387.▲
    by catigula 1y ago

    Can radiology results be instantly fed into pass/fail metrics like code can via tests? Programming is the first job AI will replace. The rest come later.
28. 388.▲
    by catigula 1y ago

    Hard to imagine that this argument exists, the real victims of mass murder aren't the actual victims of mass murder.
29. 389.▲
    by catigula 1y ago

    It's okay if they mass surveil and kill other people using sweeping AI systems, surely it will never happen to me.
30. 390.▲
    by catigula 1y ago

    We live in the age of AI, it takes approximately 2-3 minutes to get a condensed report on why your paper is misleading or incorrect: Control group mismatch: Treats non-profits as a clean counterfactual for for-profits, but the sectors have