Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
qarl
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
151.
▲
by
qarl
1mo ago
> One of the main purposes of LLMs is to launder responsibility No it isn't.
152.
▲
by
qarl
1mo ago
Well... if that's true, and I don't know that it is, it's not because of the agents. It's because of the money. People with money get away with crimes all the time and it has nothing to do with agents.
153.
▲
by
qarl
1mo ago
Whatever may or may not be happening with law enforcement - no one is confused about the liability. It's no different than when a company's machine cuts off a worker's finger. No one thinks "Gosh! The machine did it, n
154.
▲
by
qarl
1mo ago
I think it's more likely they want to call attention to the fact it was the result of agents, rather than shift blame. I'm pretty sure everyone knows that OpenAI is liable for the software they create and run.
155.
▲
by
qarl
1mo ago
Well... computer science really is math already. :)
156.
▲
by
qarl
1mo ago
Is this really any different than the problem in software engineering - where AI is doing the work of junior programmers and now they aren't getting the development they need? Seems the same to me. And it'll be the same in all in
157.
▲
by
qarl
1mo ago
This is, in fact, what my kid is planning. They are artistically inclined - make beautiful art - but when looking at colleges focused on law. Why? Because lawyers make bank, and they can fund their passion that way. Not the way I did it,
158.
▲
by
qarl
1mo ago
HA! Yes, too true.
159.
▲
by
qarl
1mo ago
> it feels like a superpower I say "it feels like flying." People worry that I'm too enthusiastic about AI. My obvious answer is - wouldn't you love to fly too?
160.
▲
by
qarl
1mo ago
To be fair - I think the world is in such flux right now it's kinda impossible to predict what is a good path forward. I have no idea what to tell my kid. But yes, I think many of us started with that book on BASIC making little progr
161.
▲
by
qarl
1mo ago
Honestly - and maybe it's because it's had longer to mature - the film industry is better than the game industry. Before computers, films already needed artists to create brilliant stuff - and programming for passion is very much
162.
▲
by
qarl
1mo ago
There's a reason garbage collectors get paid so well.
163.
▲
by
qarl
1mo ago
I think you're doing right by giving him AI. But get him a coding harness - that's where the real traction is right now. Claude Code or Codex. I say this because - the joy of making games is mostly in the design. The imaginatio
164.
▲
by
qarl
1mo ago
As someone who has worked in both the game industry and the film industry - tell your son this: When you are passionate about your work - it comes at a serious price. We live in capitalism and so everything has its cost. Passion in the wo
165.
▲
by
qarl
1mo ago
To reiterate - You insist that computers cannot effectively simulate brains despite: 1) there being no evidence for such, 2) it would be the first ever situation we've discovered, and 3) there are strong arguments that it's not ev
166.
▲
by
qarl
1mo ago
Sure. But they're trying evaluate how well the agents can research information on the web. If you deny them access to the web........
167.
▲
by
qarl
1mo ago
OH! Sorry - that's a different position then. Let's address it: So you think a computer can't simulate a human brain. But why do you think so? We've never found a physical system we can't simulate with a computer
168.
▲
by
qarl
1mo ago
If the board is going behind his back... then how does he know he was put on leave?
169.
▲
by
qarl
1mo ago
You already conceded that the simulation would act exactly as the real item. That is a concession that all the behavior is captured by the laws of chemistry. The only thing you doubt is that the qualia is captured. And my question is - if
170.
▲
by
qarl
1mo ago
Yeah. My first real use of LLMs was a couple years back when we had a legal fight. We never would have won without it. Even got the high school principal to apologize. That was delicious.
171.
▲
by
qarl
1mo ago
Thank you for continuing to engage. > A truly behaviorally accurate simulation, by which I mean the simulation has parts corresponding to everything in the original, is a very high bar of course. So to address your point about faking dir
172.
▲
by
qarl
1mo ago
> Prolific AI psychosis occurs when a person generates a large quantity of AI output without significantly increasing the real value of their work. In some cases, the new AI workflow may even destroy value. HA! I love how he entirely di
173.
▲
by
qarl
1mo ago
You need to start considering the possibility you are mistaken.
174.
▲
by
qarl
1mo ago
It buys attention for the issue. Many people (see other comments in this very post) refuse to believe these things. And by resigning he no longer has to feel personally guilty for what happens.
175.
▲
by
qarl
1mo ago
If their frontier LLMs are the results of distilling US LLMs - then stealing it back won't work.
176.
▲
by
qarl
1mo ago
LLMs are quick to jump to erroneous conclusions. I think we already knew that.
177.
▲
by
qarl
1mo ago
AIs are putting humans out of work. Yes. We already knew this. Are we actually surprised it's happening? I guess we are.
178.
▲
by
qarl
1mo ago
Hey - I noticed you're not replying to me. No worries - I understand that these conversations can be a bit much. I just wanted to thank you - our talk helped me develop my ideas around "emotion faking" - which I was having t
179.
▲
by
qarl
1mo ago
I feel the opposite. As an old man now, I can think back over my lifetime of programming, and I remember all the ideas I've had but had no time or energy to execute. I've landed 5 super fun ideas in the last 5 months. It feels lik
180.
▲
by
qarl
1mo ago
Yeah... you've convinced me. An reverse engineering LLM should/would never invent literal strings in the code.
More ›