I have and do look at it from that angle, my point was that the need for data collection isn't a guaranteed thing.

Even if you narrow it down to just the subset of companies for which continuous "shareholder value" is a requirement.

If it was the only avenue to profit i'd agree the need was universal, but there are other ways, some mutually exclusive.

Though i will conceded that data collection is one of the common ones.

It's absolutely fine at some stuff, provided you know enough to spot any mistakes it might make.

Just because you can do it with an LLM, doesn't mean it's the best tool for the job.

Oh yeah, im familiar, crossed is top tier.

Dropped off of my radar a while back, but I'll pick up again eventually

Greater compared to human code? Not sure about that, but I'm not disagreeing either. Greater compared to verified able programmers, sure, but in general?..

Both.

The reasons are quite hard to describe, which is why it's such a trap, but if you spend some time reviewing LLM code you'll see what I mean.

One reason is that it isn't coding for logical correctness it's coding for linguistic passability.

Internally there are mechanisms for mitigating this somewhat, but its not an actual fix so problems slip through.

I don't think I'm getting your point here. Do you mean by that, the code basically lacks focus on an end goal? Or are you talking about the fuzzyness and randomization of the output?

The latter, if you give it the exact same input in the exact same conditions, it's not guaranteed to give you the same output.

The fact that its sometimes close to the same actually makes it worse because then you can't tell at a glance what has changed.

It also isn't a simple as using a diff tool, at least for anything non-trivial, because it's variations can be in logical progression as well as language.

Meaning you need to track these differences across the whole contextual area which, if you are doing end to end generation, is the whole codebase.

As I said, there are mitigations, but they aren't fixes.

Let's assume we're skipping the ethical and moral concerns about LLM usage and just discuss the technical.

it makes an impression on me as if human code would be free of such errors

Nobody who knows anything about coding is claiming human code is error free, that's why code reviews, testing and all the other aspects of the software development lifecycle exist.

To me it sounds like nobody should ever trust AI code

Nobody should trust any code unless it can be verified that it does what is required consistently and predictably.

because there can or will be mistakes you can’t see, which is reasonably careful at best and paranoid at worst

This is a known thing, paranoia doesn't really apply here, only subjectively appropriate levels of caution.

Also it's not that they can't be seen, it's just that the effort required to spot them is greater and the likelihood to miss something is higher.

Whether or not these problems can be overcome (or mitigated) remains to be seen, but at the moment it still requires additional effort around the LLM parts, which is why hiding them is counterproductive.

At some point there is no difference anymore between “it looks fine” and “it is fine”.

This is important because it's true, but it's only true if you can verify it.

This whole issue should theoretically be negated by comprehensive acceptance criteria and testing but if that were the case we'd never have any bugs in human code either.

Personally i think the "uncanny valley code" issue is an inherent part of the way LLM's work and there is no "solution" to it, the only option is to mitigate as best we can.

I also really really dislike the non-declarative nature of generated code, which fundamentally rules it out as a reliable end to end system tool unless we can get those fully comprehensive tests up to scratch, for me at least.

Why don’t you give me an example of a better way to transmit information than through words, one that could have been applied throughout history?

Firstly : Why would i need to, i wasn't claiming there was a better option ( at the time ).

Secondly : I specifically mentioned spoken and written(on paper) words, not just words in general.

Thirdly : if i had to guess, probably some form of math stored in some long term medium like diamond or something, i'm just guessing though, not my area and still don't need to have a good example for a point i wasn't making.

Bonus : Another of my points was that why would i need to when there is an omnipotent deity around, whose job it is to do it.

My whole argument has been that i disagree with your repeated assertion that they are accurate to the original intention, easily understandable and not prone to misinterpretation (intentional and otherwise).

Also the "my morality is the only correct morality” thing, i cannot stress enough how much of a red flag that one is.

And it’s not that they didn’t read it properly, it’s that they simply don’t care and scammed people. Even if they had read it, which they probably didn’t, they wouldn’t care. Evangelical megachurches and their priests all know what’s up, they simply don’t care. Rapey, pedophilic priests knew what was up. The leaders of the Crusades, with strong financial interests, knew what was up. It’s not a “no true Scotsman”, it’s more like “if I’m raping a child and you find me and I tell you I’m a good moral man, a believer who’s afraid of God’s judgment, and you believe me, you’re mentally challenged”. And many people are, I guess.

You have to be intentionally skipping logical steps here, no way someone with your vocabulary accidentally overlooks the thousands/millions of "holy" warriors and zealots who wholeheartedly believe(d) in their interpretation of the guidelines.

• They aren't 's children so it's fine.
• If this leads them to the righteous path, it's fine.
• They aren't real people so it doesn't apply.
• It's not stealing it's fundraising for the furthering of our divine cause.
• If i kill/main/harm the infidels/blasphemers will reward me.
• if i dont kill/main/harm the infidels/blasphemers will punish me.
• I'm one of the chosen, i can do because has given me a divine mandate.

Come on now, pretending an easily provable demographic of people don't exist because it doesn't fit your narrative is the weakest of sauces.

You can do better.

OK so, benefit of the doubt.

You know that statement wasnt asking for protection right?

Parker from Leverage , even more so in Leverage: Redemption

Mental illness is never an excuse for doing shitty things.

It can be, someone having a psychotic episode ( that couldn't reasonably be prevented or mitigated ) that hurts the people around them has a legitimate excuse for the outcome.

Part of the actual definition of mental illness could broadly be interpreted as impairment or outright loss of reasoning and cognition.

It does require us to give them treatment to avoid harming others.

Agreed.

Though i'd say, provide the framework and access to treatment, but i think we mean the same thing.

It is dangerous to not broadly paint society as mentally I’ll.

That's a very subjective take, with very vague language and almost no value as a talking point without more specificity.

To be clear, i'm not expecting an essay or anything, i just can't really respond without more information about what you mean.

Look south of you. At least 30% of the US population is mentally ill, and they should all be given free treatment for this illness.

An interesting perspective, if somewhat US centric, i mostly agree.

None of which addresses my original criticism that the definition of mental illness isn't something that should be ascribed to " all 'terrorists' ", it means something relatively specific and terrorism isn't a good synonym.

Again, not what the definition of "mental illness" generally means.

Look up an actual definition or this

Can "terrorists" have mental illnesses?, sure.

Are all "terrorists" by definition mentally ill, doubtful.

Without even getting into the subjectiveness of the term "terrorist", lets take your example.

There are plenty of situations where you can end up with that point of view and not have a legitimate "mental illness", because that term means something relatively specific and isn't a good enough fit with which to broadly paint all members of a group.

Another example of why it doesn't fit is that there are plenty of people who are evil/bad/morally bankrupt (for whatever frame of reference you are using to determine such things) that shouldn't get to use mental illness as an excuse for doing shitty things.

I don’t agree with this. Operation warp speed was Trump’s project. When it comes to COVID, I feel like he was pro-vaccine since the beginning. Back then it was also the democrats who were expressing scepticism about the safety of it because they didn’t trust Trump and felt like the vaccine was rushed and not properly tested.

Fair enough, i don't agree that he was pro-vaccine, rather pro self-interest, but outcome wise i'm not sure it matters.

Assuming his stance wasn't antivax, it could be argued that he could have done much more with his platform to push for vaccine adoption, given the clear anti-vax stance a large proportion of his base had/has taken, but that's an entirely different argument.

He has said sceptical things about masks that has caused distrust and conspiracies in the MAGA population. I don’t see the need to defend him on that one.

Its less about you defending him and more about omitting a position that conflicts with the narrative of the reply you provided.

Emphasising the point that correlates with your (in general , not you specifically) narrative and omitting the point that doesn't, is a common bad faith tactic.

Perhaps that isn't what you were doing, but it could easily be interpreted that way, and that's what i think you were asking for when requesting examples.

Sure, but what I mean is that simply being downvoted doesn’t alone mean the information is incorrect and the opposite is true as well.

Sure, but it was written in such a way as to imply that downvote = "people don't like the truth".

Which is a classic bad faith stance to take.

That you weren't actually taking that stance is clear now, but not from the original text ( at least to me )

There’s comments in this tread with false info that’s being upvoted.

Agreed.

Did you mean to reply to me? Or the person above?