FriendOfDeSoto

I take solace in the fact that a lot of the output of so-called AI needs to be checked and doublechecked, rendering any time and overhead savings nonexistent. This may change but I'm clinging on.

It's too early to tell where on the "modern technology replaces humans spectrum" the advent of so-called AI falls. Are we talking about enraged workers seeing their livelihoods in danger by industrialization throwing their wooden shoes, sabot, into the machines? Hence, or so the legend goes, the word sabotage. Or are we talking about accountants and bookkeepers, whose need to exist was questioned when Excel became a thing and automated something like 60 percent of their work. They actually grew in number because they could do more sensible things now. We can at least hope it's the second scenario.

As far as the masses enjoying so-called-AI-generated music is concerned, I think of how the availability of photography changed paintings. When you couldn't just snap a picture of something, a photorealistic facsimile in oil on canvas was fantastic. It took weeks but you didn't know better. When photography became widespread, artists went banana. Picasso actually knew how to draw things correctly but you wouldn't think that seeing his later body of work. Impressionism is delivering lovely scenery without sticking to the realism of the Dutch masters. Art isn't in a vacuum, it develops around life, life includes technology - it's an unavoidability that technology influences art.

Any photograph would be amazing to the people in 1830. Wow! It's my neighbor Bob sitting on a chair. Wow! It's a picture of a thumb. I have two of those but I've never seen them like that. Wow! It's a picture of New York. I'll never be able to go there because it takes 6 weeks and costs more than my net worth. Jump to today and we've become much more discerning about what a good photograph is. I live in hope that we develop a so-called-AI discerning taste as well. Especially in music. We've done okay with photoshopped images too.

I know. I'm a creature of habit.

"Greedy CEO" seems tautologic to me;)

I hear you. I just would like to point out von Ahn wasn't talking about today. He's looking at a future where some of these tools actually deliver on their promises. I understand why you're skeptical and frankly so am I. But there is a chance he might prove us wrong in our lifetimes.

He's probably right that a so-called-AI software can assist a teacher in making sure all students get to be on the same page. With the unreliability of models today though I would be more concerned with the crap hallucinating wrong math formulae or the Italian-Zimbabwean War of 1647. This needs tight supervision by professionals. But in his defense, he was just shooting the breeze and didn't give a time frame. In a decade this scenario might look less wacko. But we also thought we'd be fizzing around in flying cars already.

What else is he gonna say though? They pivoted hard into it, of course he's gonna sing the praises. In other news, water is wet.

In my experience, the courses on DL are getting worse. Erratic changes, mistakes, etc. I'm just not ready to kiss my 1000 day streak goodbye yet!

When you're using lemmy or mastodon, you don't have to use the website. You can use an app that goes from your fingers to the server without needing a browser and a website to exchange the information.

So most if not all the instances of the fediverse are also a website if you need to use it. But not every website is an instance of the fediverse.

But you had Facebook. That's as good as having it. They know you. Their grubby tentacles will never let go!

Whether you like it or not, they probably already know who you are too. They're collecting shadow profiles of people who haven't signed up through various means.

If others have posted pictures of you on a meta service, there is a good chance it already knows what you look like and they know it's you even if you're not tagged.

People who allow them access are just less work for them. And now they have info to train their so-called AI models. Now it's a question about what are they going to with them. The application is wide. Create fake pictures, create fake profiles, etc. And at some point we will find out about a massive data leak that happened because the company is run by unapologetic sociopaths.

Klingelingeling!

Socialism badge unlocked.

I think there may be a paradox hiding in your question. You cannot believe in free will. You have it or you don't - I would postulate you need a neutral third-party observer to tell you. For us humans, a Martian might do. Believing is an act of faith. Faith tends to bend will to its dogmas. I would go so far as to say belief is the natural enemy of a free will.

We are distracted animals. All things being equal, the Martian observer will after years of careful study come to the conclusion that humans have free will. But it's constantly battered by short attention spans, a tendency to go with the herd, presupposituons in our heads that we don't often or never question, etc. We are a smartphone full of bloatware running on too little RAM. It takes skill to operate. Some are more skillful than others.

You could of course counter that by saying that's what you believe. It's paradoxes all the way down.

In their defense, they were probably lying before the advent of so-called AI as well.

If we take the forum title here, the "fuck" is directed at the people in charge of so-called "AI" companies. The technology has value. It's just being forcefed down our throats in ways that remind us of block chain and whatever happened to block chain?!