AI is like a hammer

ironhydroxide@sh.itjust.works to Showerthoughts@lemmy.world – 122 points –

Any tool can be a hammer if you use it wrong enough.

A good hammer is designed to be a hammer and only used like a hammer.

If you have a fancy new hammer, everything looks like a nail.

53

Some hammers use enough energy to power a small country in order to show you a cake recipe without an entire backstory and 50 ads.

A cake recipe that instructs you to put non-toxic glue in it, and some small pebbles on top of it.

Welcome to capitalism.

AI is the new thing, so cramming it into a product increase funding and/or stock price.

Even if it hurts the product.

Even if it hurts the product

Because the product is not the product. The stock valuation is the product.

It's more like Jefferson's dumbwaiter, in that it was created by someone who verbally supported an egalitarian utopian vision of society, but the device itself is a scale model of an exploitative social system. At one station of the device, unpaid/low-paid labor operates out of view of the user, and then at the other station, the user enjoys an almost-magical appearance of an answer to their request.

No tool is "just a tool", after all. In that way, AI is like a hammer.

(That section of the video leans heavily on Do Artifacts Have Politics?, which is a pretty short and accessible essay. If you're not convinced that artifacts do have politics, and you don't want to watch the video, just read a few paragraphs of the essay.)

Nope. It's more like that weird thing you brought at 3 am off of the Home Shopping Network because you were in a really bad place and thought it would make you feel better.

Now it's taking up space and you don't want to throw it out because that would mean you're a failure...

It's not good at replacing your job, but good at convincing your boss that it can

Basically what I said to people who asked me about my opinion on AI.

Exactly it was: "AI is a tool like a hammer. If you hit your finger, don't complain about the tool, but because you simply used it wrong."

Except AI is a pipe wrench pretending to be a hammer.

True.

And I get where you're going, but pipe wrenches are still way too useful in too many situations. AI is a like a disc brake compressor hand tool, being sold as the solution to everything else.

When I mention how much I like it for compressing a disc brake, I feel like people look at me like I'm crazy for falling off the hype train.

Edit: And by people, I mean AI hype shill bots, probably.

Sometimes it is more like "AI is like a hammer in a world full of screws."

calling (mm)LLMs AI is just corpo bullshit. But hey, it's fancy, right?

I don't see the "is not actual AI" argument.

Since the 80 AI has just been algorithms and proposals for neural networks.

It never has need to have a "soul" or "be sentient" to be Artificial Intelligence.

Even a simple Tic Tac Toc opponent algorithm has been called AI without much complaining about it.

Also AI didn't got called AI by corporations. That naming for the technology dates from where it was being proposed as concepts in universities.

you not liking it doesn't make it any less ai. I don't remember that many people complaining when we called the code controlling video game characters ai.

Or called our mobile phones "cell phones", despite not being organic. Tsk.

Except cell phones or cellular phones refer to the structure a mobile network is built on: a mesh of cell towers.

pretty sure that they were and still are called Bots though, atleast in the context of first person shooter.

Software developer, here.

It's not actually AI. A large language model is essentially autocomplete on steroids. Very useful in some contexts, but it doesn't "learn" the way a neural network can. When you're feeding corrections into, say, ChatGPT, you're making small, temporary, cached adjustments to its data model, but you're not actually teaching it anything, because by its nature, it can't learn.

I'm not trying to diss LLMs, by the way. Like I said, they can be very useful in some contexts. I use Copilot to assist with coding, for example. Don't want to write a bunch of boilerplate code? Copilot is excellent for speeding that process up.

LLMs are part of AI, which is a fairly large research domain of math/info, including machine learning among other. God, even linear regression can be classified as AI : that term is reeeally large

I mean, I guess the way people use the term "AI" these days, sure, but we're really beating all specificity out of the term.

This is a domain research domain that contain statistic methods and knowledge modeling among other. That's not new, but the fact that this is marketed like that everywhere is new

AI is really not a specific term. You may refer as global AI, and I suspect that's what you refer to when you say AI?

it's always been this broad, and that's a good thing. if you want to talk about AGI then say AGI.

I know that they're "autocorrect on steroids" and what that means, I don't see how that makes it any less ai. I'm not saying that LLMs have that magic sauce that is needed to be considered truly "intelligent", I'm saying that ai doesn't need any magic sauce to be ai. the code controlling bats in Minecraft is called ai, and no one complained about that.

10 more...

They are neural networks which are some of the oldest AI tech we have.

You can hate them, but they are by definition AI.

uhm no, AI is relaying on neural networks (which are just weighted gates) but Neural Networks are not by definition AI.

Complaining that it's called AI is like complaining that smartphones are called smart. There's no stopping it, you just end up sounding like an old man yelling at the cloud. (Which isn't really a cloud, but we still call it that)

Nah, smartphones being actually "smarter" than feature phones as in you can do way more than just basic stuff like calling, messaging people, run a simple calculation, having a calendar etc.

10 more...

I am really piss off when Reddit use AI to shadowbanned my account. They never tell the reason. They just hide my interaction to outerworld, assuming that i am dumb and never found out.

As a result, all subsequent accounts i try to create are shadowbanned after 5 minutes because my phone in blacklist.

Appeals everyday but never get responded. I am sure all of this is handling by AI.

I wouldn't even call this AI, they just have an algorithm. That's poorly tuned.

I encountered the same issue, and that's exactly why I'm here on Lemmy. Reddit and it's Shadow banning habits have to go away.

I don't like having persistent social accounts, so I make a new one for each topic, Reddit, GitHub. Purpose specific accounts to do one thing. And for the last few years, every time I create an account like that, it's immediately shadow banned. It's frustrating, because my contributions are now thrown away, and it's dishonest, because these services don't have the politeness to even tell you you're not allowed to participate.

Because of that, a federated system like Lemmy must survive. That's why we're here

You wouldn't call it AI. But you can bet the C suite sure calls it AI.

Conways game of life is ai :)

You're right! If Conway's game of life wasn't AI a year ago, it sure as hell is now!

I wonder how many rubes I can sell it to...?

First make it Conway's as a service, then get at least a $3 billion dollar valuation, get your seed round in. And then sell it before you actually have to deliver any revenue.

It's perfect