martin-t

46 ms·

631.
by martin-t 1y ago
> I tell it the secret simple thing it’s missing and it gets it. Anthropomorphizing LLMs is not helpful. It doesn't get anything, you just gave it new tokens, ones which are more closely correlated with the correct answer. It also g
632.
by martin-t 1y ago
That's incorrect, gambling is about waiting. Brain scans have revealed that waiting for a potential win stimulates the same areas as the win itself. That's the "appeal" of gambling. Your brain literally feels like it&#x2
633.
by martin-t 1y ago
That was a figure of speech. It should probably be something like `max("harm caused adjusted by level of intent", "harm intended")`.
634.
by martin-t 1y ago
The issue with many laws is that they take into account the individual case. If somebody has a track record of lying, it should be easier to prove subsequent lying. Of course, then the potential for abuse is greater. It's about finding
635.
by martin-t 1y ago
> won't get read > responds to said wall of text Saying any more is a waste of time. Bye.
636.
by martin-t 2y ago
Your argument contains multiple fallacies. You first act as if the current situation is the best we can do by pretending that no alternative can be better by implying that any alternative is naive. I attempt to be reasonable and explain in
637.
by martin-t 2y ago
Interesting, I didn't know there was so much disagreement about using it.
638.
by martin-t 2y ago
I've read this too but it doesn't disprove what US was thinking at the time. People think others think like them. US being a democratic country and considering the value of a life to be high, I have no trouble believing that the U
639.
by martin-t 2y ago
Lying by omission is still lying. What I've noticed is that for a lot of people, if you do something wrong through a sufficient number of steps, they feel like the severity is lower. The opposite is in fact true - causing harm through
640.
by martin-t 2y ago
"A bad plan now is better than a good plan later." People have evolved to unify behind a strong (and aggressive) leader because historically the biggest threat to one's tribe (and therefore genes) were other tribes. You might
641.
by martin-t 2y ago
Obviously all available tools will by used by bad people. What we need is: 1) Good people to also use those tools - a lot of self-proclaimed good people think some tools are bad and therefore they won't use them. But tools are just too
642.
by martin-t 2y ago
I never said we didn't need rules, just that when they are too specific, people tend to follow the letter but break the spirit of the rule. (Sidenote, one deeply ingrained idea is that the law is somehow special compared to other rules
643.
by martin-t 2y ago
There are glimmers of hope - like Wales trying to ban lying in politics. But of course, the punishment has to be proportional to the offense, not just a slap on the wrist. If I wanted to take things to an extreme, I'd ask why laws even
644.
by martin-t 2y ago
> By that logic, we should skip the depleted uranium and head straight to thermonuclear weapons Yes, actually. (With a massive caveat being if the opponent does not also have nukes.) I mean, why do you think the US nuked Japan at the end
645.
by martin-t 2y ago
A do believe causing harm without justification should automatically result in punishment that causes the same harm to the abuser multiplied by a multiplicative constant but 10x is probably too much. Usually, I'd suggest something betw
646.
by martin-t 2y ago
This is the kind of implicit lying that seems pervasive today and I am so tired of it. This alone is sufficient evidence of their malicious intent and should be enough to punish the people responsible for trying to ruin an innocent person&#
647.
by martin-t 2y ago
I wish people would seriously consider (A)GPL for their projects more often. It hasn't happened here, though has certainly happened in the past without anyone knowing - (A)GPL would make it hard for them to make a closed source "f
648.
by martin-t 2y ago
> Interesting that "psychotic disorders" can be replaced with "tribalism" and still be true That's because both are defects of rational reasoning. I view it this way: The only difference is "disorders"
649.
by martin-t 2y ago
For me the biggest indicator of how much bullshit A"I" is, is how the A"I" tools themselves are bad. Visual Studio still doesn't have an "accept one suggested word" shortcut. If your LLM is so great, why d
650.
by martin-t 2y ago
I don't even see why "hackers" and "founders" should be related in any way. The hacker attitude (or approach to life) is that of curiosity and the need to build and create. The founder attitude AFAICT is how to get
651.
by martin-t 2y ago
Oops, the one time I don't check, I end up getting it wrong, sorry.
652.
by martin-t 2y ago
Reviewing code is harder than writing it. When _writing code_, you achieve a certain level of understanding because you fundamentally need to make all the decisions yourself (and some of them are informed by factual statements). When _readi
653.
by martin-t 2y ago
"Can" add in the chance if nonzero? Sure. "Easily"? No. Just check the article, it's plenty of examples of skilled programmers stopping to use A"I" because long term they discover more bugs than they would
654.
by martin-t 2y ago
Whenever i open HN, there's a pro-A"I" post which makes me almost question if intellectual labor will be useful in a few years. Yet when i try to use these tools, they fall short. They can probably be useful for generating ha
655.
by martin-t 2y ago
This has been on my mind ever since I realized 2 things: - the difference between zero-sum and positive-sum games - that large parts of society are engaged in zero- (or even negative-) sum games 1) some through choice or 2) because they are
656.
by martin-t 2y ago
The pressure is simple - money. Competent people are rare and we're not cheap. But it turns out, those cheaper less competent people can't replace us, no matter what tools you give them - there is fundamental complexity to the wor
657.
by martin-t 2y ago
Not necessarily confusing but very frustrating. This is probably the first time I encountered such a wide range of opinions and therefore such a wide range of uncertainty in a topic close to me. When a bunch of people very loudly and confid
658.
by martin-t 2y ago
Whenever I read about LLMs or try to use them, I feel like I am asleep in a dream where two contradicting things can be true at the same time. On one hand, you have people claiming "AI" can now do SWE tasks which take humans 30 mi
659.
by martin-t 2y ago
It literally says "for Rust applications" on its front page. Your tone is insulting and has no place in technical discussions.
660.
by martin-t 2y ago
Niko Matsakis made a proposal on his blog for opt-in contexts with relaxed rules (such as implicit conversions) and it was hated, at least on reddit. I doubt he would delete it but i can't find it now.

More ›