[flagged]
Not just programming. e.g. You have a complex medical problem. Hard to ask Google. Ask AI, it gives you some possible answers, then you can search up those. Or you want to identify a plant. AI looks at your photo and tells you what it is, then search that name to verify.
There were existing ways to do some of these things, but this covers all of them.
In essence, they are only adequate in niche situations (like creative writing, marketing, placeholder during iterative design, …) where there's no such social contract and assumption that people operate in good faith and do their best diligence not to deceive others.
Pretending otherwise, not pushing back when LLMs are clearly used outside of those contexts, or dressing them into what they are not (thinking machines, search engines, knowledge archives, …) is doing the work of useful idiots defending tech oligarchs and data thieves against their own interests.
And yeah, I get it, naysayers are annoying. Doesn't mean they are wrong or their voices shouldn't be heard at a time where the legality and ethics of all this are being debated.
Placeholders, sure. But I'm surprised you don't know about the utility it has in a programming context.
I was kicking the tires on Claude for a personal thing the other day - "given this data and this output I want, write a utility function" - and it quickly gave me something that gave me exactly what I needed on the first try.
I also had an extended session recently where it really helped me with the "naming is hard" problem. Naming is hard and it's nice to have a rubber duck that has a certain understanding of your code and is able to throw a bunch of ideas at you quickly.
I could go on, but hopefully you get the point.
And I'll emphasize your "placeholder" point: sometimes it's not even about the solution it gives, it's about the process of articulating the problem and getting something out there. A lot of times the first iteration of anything is the hardest to complete, and it can really cut down the time to achieve that.
It is in that sense that I see LLMs challenge "social contracts": their use is infectious (probably also in the GPL-sense, be warned) and should be agreed upon by all parties upfront (do you want to be on the receiving end of a software that you pay good money to have developed while knowing that its developers have little understanding of its workings? How many times cheaper is cheap enough to pay-off the amount of technical and institutional debt/risk? Can LLMs effectively become that cheap?).
What does that mean? Are you implying that I used a LLM to formulate my comment? A comment about the dangers of LLMs for civic and respectful discourse?
> Ultimately saying a LLM is only useful in niche situations
is a fact. LLMs are not something new or obscure, we have plenty of open-source/weights implementations keeping in check the outlandish marketing promises of the big players and their "SOTA" models.
> …is a tell of the boat passing you by.
is an opinion, giving an emotional twist to something that really doesn't deserve one. All it shows is that you bought into LLMs being more than what they really are, and welcome the (admittedly cool and tempting) marketing above the truth and your critical and independent thinking on the subject.
We can debate usefulness all day, but reducing the opposing view to “you fell for marketing” isn’t critical thinking.
And yes your reply unfortunately read exactly like ChatGPT would respond which I found both amusing but also hard to read as the logic was overly verbose.
I'm sorry but I can't assume that your comment comes in good faith. You are the one making the exceptional claims here, not me. The burden of the proof falls on you.
From my side, I've provided clear use-cases where LLMs are adequate, and large categories of use-cases where they are provably not (misguided usage in those by a number of people is no counter-argument).
The extent of your argumentation is annoyingly nihilist "we don't know everything about LLMs, so I'm entitled to believe anything I want".
> We can debate usefulness all day, but reducing the opposing view to “you fell for marketing” isn’t critical thinking.
The reason why I'm pulling marketing into this is because that's the reason why people generally believe that LLMs are more capable than they really are. Placarding LLMs everywhere and over-selling their abilities is a coordinated effort by a very few immensely rich tech companies. Not by Machine Learning scientists or Computational Neuroscientists who have been mapping this same space for decades.
> And yes your reply unfortunately read exactly like ChatGPT would respond
Then I suppose it's a compliment, taken as someone whose native language obviously isn't English.
Nice chatting, but it’s clear you’re out of productive things to say.
I mean, one easy thing you could do is to explain how LLMs are adequate as
>> thinking machines, search engines, knowledge archives, …
How much practical experience do you have with LLMs? Have you even tried running models locally? Then how have you not experienced yourself tweaking few parameters and gotten the same model to say one thing and its opposite just after? Through your extensive use of commercial LLMs, aren't you getting daily occurrences of confabulations? Making stuff up is what defines¹ LLMs, and that's not me saying it.
If your field is so inconsequential that reliability and faithfulness of your output has no practical merit or repercussion, I'm glad that you get some entertainment out of this. Still doesn't make it the norm, though.
Don’t misjudge me either, I am not suggesting there should not be critical comments but that if you make a critical comment you should at least have an idea what your talking about.
* give an error
* return the wrong result
* not be internally consistent with the rest of the content
* be logically impossible
* be factually impossible
* have basic errors
It is entirely possible (and quite common) to know something is wrong without knowing what a right answer is.
How long do your teams take to write vs review PRs? How long does it take to review a test case and run it vs write the implementation under test? Or to verify that a fix a regressed test now completes successfully? How long does it take you to do a "design review" of a rendered webpage vs to create a static webpage? How long does it take to evaluate a performance optimization vs write it?
> How long does it take to review a test case and run it vs write the implementation under test?
If you blindly trust a passing test and don't review it as production code, I have a bridge to sell you.
> How long does it take to evaluate a performance optimization vs write it?
Factoring in the time to review that the optimization didn't introduce a regression, and isn't a hack that will cause other issues later: the difference shouldn't be too large.
Yes, code usually takes more time and effort to write, but if it's not thoroughly read, understood, and reviewed, it can cause havoc someone will have to deal with later.
This idea that just because LLMs help you write code quicker will make you or the team more productive is delusional. It's just kicking the can down the road. You can ignore it, but sooner or later someone will have to handle it. And you better hope that it happens before it impacts your users.
an example of things that are the opposite is "public policy development", which is why it's simply malicious that various corrupt oligarchs are pushing for it to be used for such things.
so, a simple model for you to understand why other people might find these tools useful for some things:
- low stakes - doesn't matter that much if the output isn't Top Quality, either because it's easy to fix or it just doesn't matter
- enormous gap in cost between generation and review - e.g. coding
- review systems exist and are used - I don't care very much if my coworkers use an LLM to write code or not, since all the code gets reviewed by someone else, and if the proposer of the change doesn't even bother to check it themselves then they pay the social cost for it
If your quality threshold is so low that you can tolerate crashes on invalid input, you will certainly cut corners when building software for others. I wouldn't want to use a piece of software you wrote, let alone have you on my team.
> I don't care very much if my coworkers use an LLM to write code or not, since all the code gets reviewed by someone else
Ah, yes, let's kick the can down the road.
> if the proposer of the change doesn't even bother to check it themselves then they pay the social cost for it
... The side effects of shoddy code are not redeemed by "paying a social cost". They negatively impact your users, and thus the bottom line of your company.
why? my shitty sales dashboard at work doesn't control a rocket or a pacemaker. the crappy my-weird-org-mode-table-to-re-arranged-CSV convertor isn't either.
not all software is safety critical, and in any case, I'm the human who ran the code generator and then the code and I'm responsible for my dashboard crashing or my convertor deleting the photos of my cat.
should an LLM replace human code review? no. can I use an LLM for my own dumb projects? of course.