AI is a tool. And like any tool its introduction creates proponents and enemies.

Some might say I'm a semi-professional writer. As in: I make money with things I write. From that perspective, I see both the AI slop and the benefits. I love that AI gives me an on-demand proof-reader. I don't expect it to be anywhere near a professional in that field. But if I want to quickly check a text I wrote for specific things, AI is great, because unlike me it hasn't been over that sentence 20 times already and still parses it completely.

As for AI writing - for the moment it's still pretty obvious, and it's mostly low-quality (unless some human has added their own editing).

The same way that the car, the computer, e-mail and thousands of other innovations have made some jobs obsolete, some jobs easier, and some jobs completely new, I don't see AI as a threat. And definitely not to my writing. Though good luck Amazon with the flood of AI-written garbage now clogging up your print-on-demand service.

Right.

Yes, I guess it will. By hopefully making people switch away from YT music en mass.

Trivially obviously not. The LLM wasn't trained on texts exclusively written by the human using it, so it won't ever speak like that particular person.

If someone wants to train a specific "Tarrof" LLM - go ahead. I'm simply advocating against poisoning the already volatile generic LLM data with more human bullshit.

This post was indeed brought to you by ChatGPT.

That is true but also besides the point. Communicating like "a human" is the point here. WHICH human, exactly? We already have problems with hallucinations. If we now train them on huge data sets intentionally designed for the human habit of saying the opposite of what you mean, we're adding another layer of problems. Maybe get the other ones solved first?

Of course. This is a forum, not a scientific conference, so I'm not speaking in the most precise language.

It doesn't change the point.

Make piracy the better alternative in every way and people will pirate.

Make a legal transaction the better alternative in every way (ok, except money) and people will use that.

It really is so simple, even an executive should be able to understand it.

While that might be an interesting technical challenge, one has to wonder why. Just because something is "culture" doesn't mean it should be copied. Slavery was part of human culture for countless millenia. To the point where we haven't even gotten around to updating our "holy books" that all treat it as something perfectly normal. That's how normal slavery used to be.

(for the braindead: No, I'm not comparing Taarof to slavery. I'm just making a point with an extreme example.)

The thing is something called unintended consequences. So in order to teach an LLM Taarof you have to teach it to lie, to say things that don't mean what the words mean. And to hear something different from what the user says. Our current level of AI already has enough problems as it is. Do we really want to teach it to lie and to misread? Just because some people made that part of their culture?

Instead of treating LLMs like humans, how about just treating them as the machines they are? I'm pretty sure the Persians don't expect their light switches to haggle over whether to turn on the light or not, right? I stand corrected if light switches in Iran only turn on after toggling them at least three times, but I don't think so. In other words: This cultural expectation only extends to humans. Maybe just let the people complaining know that AIs are not actually human?

It holds up in some cases, not in others, and calculating an average muddles that.

Personally, I use AI coding assists for two purposes quite successfully: a) more intelligent auto-complete and b) writing a piece of code using a common, well understood algorithm (i.e. lots of sources the AI could learn from) in the specific programming language or setup that I need.

It turns out that it is much faster and almost as reliable to have the AI do that then finding a few examples on github and stackoverflow, checking which ones are actually decent, and translating them myself.

Anything more complex than that and it starts being a coin toss. Sometimes it works, sometimes it's a waste of time. So I've stopped doing that because coding it myself is faster and the result better than babysitting an AI.

And when you need to optimize for a specific parameter - speed, memory, etc. - you can just about forget AI.

Hey, industry, I've got an idea: If you need specific, recent, skills (especially in the framework-of-the-month class), how about you train people in them?

That used to be the norm. Companies would hire apprentices, train them in the exact skills needed, then at the end hire them as proper employees. These days, though, the training part is outsourced to the education system. And that's just dumb in so many ways.

Universities should not train the flavour of the moment. Because by the time people graduate, that may have already shifted elsewhere. Universities train the basics and the thinking needed to grow into nearby fields. Yes, thinking is a skill that can be trained.

Case in point: When I was in university, there was one short course on cybersecurity. And yet that's been my profession for over two decades now. There were zero courses on AI. And yet there are whitepapers on AI with me as a co-author. And of the seven programming languages I learnt in university, I haven't used even one of them ever professionally and only one privately (C, of course. You can never go wrong learning C. If you have a university diploma in computer science and they didn't teach you C, demand your money back). Ok, if you count SQL as a programming language, it's eight and I did use that professionally a few times. But I consider none of them a waste of time. Ok, Haskell maybe. The actual skill acquired was "programming", not a particular language.

Should universities teach about AI? Yes, I think so. Should they teach how to prompt engineer for ChatGPT 4? Totally not. That'll be obsolete before they even graduate.

So if your company needs people who have a specific AI-related skill (like prompt engineering) and know a specific AI tool or model - find them or train them. Don't demand that other people train them for you.

FFS, we complain about freeloaders everywhere, but the industry has become a cesspool of freeloaders these days.

The number of people isn't the problem here.

The "started every" is.

How did they not catch that during development and found a solution? I mean, the meme's where a TV ad starts Alexa and orders 10 large pizzas are a decade old now.

When you look at all of Europe, it's more like a weekly thing, and sometimes a daily thing. The ones that make the news are the bigger bombs like the thousand-pound bomb mentioned in TFS or the rare ones that cannot be made safe and have to be detonated in place, which can mean a lot of new business for window installers even with dampening.

You can't be the "baddest kick ass person on the block" without having an effective ability to fight. That means weapons, and it means training on how to use them most effectively. When going up against the Soviet Union, that means a nuclear arsenal. We made plenty of mistakes along the way, but we also helped ensure through deterrence that the Soviets never moved on Western Europe, and we helped ensure through diplomacy that World War III never broke out.

I still have my very first credit card. It's 20+ years old, sits in a lockbox, and has a thousand dollar limit, by far the lowest of my cards. It costs me $10 a year to maintain. It's there entirely because it anchors the oldest account ranking for me.