Hacker Newsnew | past | comments | ask | show | jobs | submit | ChrisLTD's commentslogin

> AI agents are just software, after all. I wasn’t going to understand them by reading prompt guides. I had to use them for real work, see where they failed, change the prompts, tools, or environment, and try again.

AI agents are just software, so the only way to understand them is trial and error?


Building understanding through use, as opposed to reading, is not a controversial idea.

Tried and true approach. Sometimes the only way to proceed if you are dealing with - embedded devices, protocols, long and deep stacktraces or freecad.

I think that's similarly in line with software; you don't really know the limitations of a library until you use it, and see that it fails for X reason at 10Hz, or 1000Hz.

I'm not sure how you can be confident the things "you" built won't crash if you're not at least scanning the code that you're generating.

human written code also has bugs and crashes…

Then why be confident the LLM code won't crash

Some people like doing DIY plumbing and related tasks around their own homes. Work might not be "fun" in the traditional sense, but it can be satisfying to build and fix things.

That’s also true of coding. There’s nothing that sucks the fun of it as effectively as doing it for pay.

Would you say that the government telling those people that they should be afraid of vaccines will help or hurt in this case?


I gotta be honest, I don't really think these immigrant groups see or care what Trump/RFK are saying. If anything they see them as an "enemy" who they should not trust.. This is an incredibly left-leaning city where the majority of public health information (or any other messaging) being disseminated is counter to what the current federal admin is saying (as a rule, it seems).

So to answer your question, I don't personally think it helps or hurts in any significant amount relative to other cultural drivers.


I don't think people base their beliefs purely based on what the administration says

the whole rollout and lockdowns, Fauci, etc. were a perfect storm to cause skepticism towards vaccines


I don't think it was the roll-out or anything along those lines, so much as the lying. Saying people won't get sick or won't catch COVID if they get the vaccines [1] was just false and very myopic. Obviously the intent was to try to get people to take the shots, but they completely destroyed trust in the process. Broader skepticism predictably follows from there.

[1] - https://xcancel.com/ITGuy1959/status/1581034815700488192


> Yet, we all trust these things.

We trust them - in large part - because we know that professionals took the care and responsibility to build the foundations correctly.

And if the foundations failed in some way, those same professionals would suffer a hit to their reputation.


> upper management who are already eliminating these assembly line, JIRA-ticket-taker software jobs en masse

Is there any proof of this?


Some anecdotal trends listed in this report on US demand for Indian tech workers rapidly slowing.

https://thefederal.com/category/news/h1b-visa-indian-tech-wo...

> According to the discussion, foreign hiring at Google has fallen by more than half, while approvals at Amazon have dropped by nearly a third.

> According to Xfino's Active Tech Jobs Outlook, active technology job openings fell to 93,000 in June, down 14 per cent from 108,000 a month earlier.


> I fully admit that we have been surpassed by the machine in the ability to entertain.

Why though? Daily here on HN, posts are pilloried for being AI generated and full of AI tropes.

AI generated content might be drowning out human generated content, but I wouldn’t say that it’s surpassed human content in a qualitative way.


I think in order to notice and take offense to those tropes, you already have to be prejudiced against them. And you probably also have the pretense that what you're reading was thought important enough by another person to write. So when you realize that this isn't the case you become disappointed.

But some people just don't care, and are just in it for novel rearrangements of the narrative structure. I'll give you that I don't notice this so much with literary content, but I have seen people being entertained by AI generated music and videos quite a bit, especially children and those who in general are uncritical of what they consume.

For example, the other day I was at a restaurant where they were playing some AI abomination of Frank Sinatra called "Take me to the Moon", and nobody really seemed to notice or care while I was becoming nauseous to my stomach. You have to already be biased to notice it, let alone to even be bothered by it.


Most popular fiction is formula entertainment like Tom Clancy novels. Tom Clancy novels are their own industry now. I have zero doubt that AI tools can generate countless variations of good guy US special OPs going after bad guy terrorist books. The "AI tropes" are over sold on HN. With the correct prompting and iteration today's AI tools can make nuanced fiction which avoids the low effort tropes. BTW "high effort" work with AI tools is but a small fraction of what of the effort using traditional means.


This is certainly overstating the difference between compiled code and the source code. If I write an if/else statement, I understand what the computer is going to do when it's evaluated even if I don't know or see the literal machine code.


Only if the compiler is perfectly naive. With the state of today's optimizing compilers, the output might be very different from what you'd expect. But it also doesn't matter because the business goal is the results from executing the program, not the static code. Hence why, as the earlier commenter pointed out, efforts go into validation rather than understanding the generated code. 99% of the time the code a compiler spits out simply doesn't matter. The same is now proving true as we move up a level of abstraction. The code spit out from an LLM doesn't matter any more than the code spit out from a compiler, only validation that it functions as expected matters.


compiler is not putting words together randomly with a mathematical probability but the LLM is


Yes, the difference between a compiler and an LLM is that a compiler's output correctness is binary, whereas an LLM's output correctness is probabilistic.

That may be an interesting distinction in a vacuum, but given that the input to both is words put together randomly with a mathematical probability a compiler's binary correctness isn't actually the useful property that it may first seem. In the real world, to overcome the probabilistic input, you have to validate the input with respect to the output. And it turns out once you have systems in place for that, you gain the same assurances with a probabilistic output as you do with binary output.

So, while there are some technical differences, there is no practical difference.


There are other benchmarks for that


The datacenter antipathy is odd, but then again so are the AI corps' marketing strategy of making everyone afraid for their jobs and the future of the species.


What else would they advertise with? "Our model constantly makes mistakes so you still have to hire expensive humans to proof-read everything"? The entire point of the AI industry is to automate human jobs. There is nothing else to advertise with.


"Our AI makes your workers X times more productive"

"Our AI takes the drudgery out of X, Y, and Z"


> "Our AI makes your workers X times more productive"

This is still replacement, it's just not _total_ replacement.


Whatever Elon Musk, Jeff Bezos and Peter Thiel are calling for, the public instinctually wants the opposite.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: