Technology

There can be nothing new or original out of AI because all of its inputs are stolen from what already exists. Real creativity comes solely from humans. Also, that clip - the song, singing, and visual - is dreadful in every way.

This needs to be hammered into techbro's heads until they shut the fuck up about the so-called "AI" revolution.

How to make line assembled pop music even more sterile and devoid of soul. Now at one thousandth the cost! But wait, how are CEOs going to molest child artists when they aren't real?!

That's pretty amazing.

The song sucks, but here was the cutting edge of AI music just seven years ago.

That it's gone from some nightmarish fever dream mashup to wannabe pop influencer levels of quality in less than a decade is pretty crazy, and as long as there isn't a plateau in the next seven years we'll probably be in a world where AI generated musical artists have a popular enough following that they will have successful holographic concert performances by 2030.

I over and over see people making the mistake of evaluating the future of AI based on the present state while ignoring the rate of change between the past and present.

Yeah, most of your experiences of AI in various use cases is mediocre right now. But what we have today in most areas of AI was literally thought to be impossible or very far out just a number of years ago. The fact you have any direct experiences of AI in the early 2020s is fucking insane and beyond anyone's expectations a decade earlier. And the rate of continued improvement is staggering. Probably the fastest moving field I've ever witnessed.

The fact that AI can produce this is impressive as to where we have come with AI. But can this actually threaten human artists?

In the United States, a federal judge ruled in 2023 that AI artwork cannot meet federal copyright standards because “Copyright law is ‘limited to the original intellectual conceptions of the author’.” With no author, there is no copyright.

~~https://www.makeuseof.com/copyright-rules-ai-art/~~ See u/[email protected] 's article below.

"The answer will depend on the circumstances, particularly how the AI tool operates and how it was used to create the final work," the office said.

Under current US law, that song is probably now in the public domain. If the law changes, that could mean that in the future, music charts potentially could be filled with AI songs. As it stands, this is most-likely a public domain music machine cranking out music that anyone can use royalty-free. It depends on the interpretation of the courts.

Lots of hostility, but that's to be expected. Research has shown that humans generally like AI-generated content as long as they don't know it's AI generated content, and when they find out they switch over to hating it.

It's simply bias. It'll go away eventually.

Looking at the hateful comments gives me shivers when thinking how humans will proceed with machines on an emotional level.

If we ever reach sentinent AI, it will go towards I-Robot plot. Ill bet.

Edit typos

I have a feeling they knew how this would be received considering it seems like they're rage-baiting and acting pretentious to try and get attention.

It's not as though the article is any better - bots just wrapping up a comment thread from twitter, cramming it with ads, and presenting it as an article.

Late stage capitalist media is a dumpster fire.

That picture is weird, there's some AI nonsense going on with the microphone shock mount, and her jaw doesn't line up with the rest of her face. Plus the usual uncanny valley weirdness of an AI generated image.

Not even going to bother with the song.

Not so much an expert rather than just parroting what I saw, but according to a cursory search, and a Twitter community note I saw from a friend, the voice used by this project is Synthesizer V, which is actually a perfectly legitimate piece of software for using digital voices in music production. If you've heard of Vocaloid, or know about Hatsune Miku, SynthV is basically a competitor in that space.

Going back to the community note, the voice used is actually called Natalie, and apparently the TOS of SynthV does not allow use of its voices using a name that's different to what was given. So they essentially can't present the Natalie voice as Anna, which they are.

EDIT: I want to clarify that these voice synthesisers like SynthV and Vocaloid are usually based on the recording of someone who has consented to the use of their voice in that regard and has been paid for it. It's not like the current AI voice cloning trend going on.

This is the worst AI will ever be, again.

Right now, at the end of 2023, you are seeing barely a year of public interest and widespread development, after maybe a decade of slowly grinding academic experimentation. And already it's enough to build some Vocaloid knockoff from scratch. You can tell it's fake, as surely as a seven-fingered hand on some anime girl staring dead into the camera. But if you think all AI drawings still look like that... you should go check.

This isn't a threat to artists, though. It's a threat to the industry. Real human beings who want to make art will have more and better tools than ever before. Audiences that want an endless spigot of AI content... won't need recording studios. You can already run this stuff on your computer. Some networks are getting better by getting bigger, which demands a really fancy computer. Other networks are getting better by getting smaller. Smaller networks train faster, even if they're deeper, more abstract, and less predictable. They run faster, too, and on lesser hardware.

Hold onto your butts, folks. It's gonna get weird.

Also, far from the most pressing issue here, but: just say Twitter. You don't have to respect the stupid rebrand. You know it's stupid because everyone keeps clarifying what they mean.

Here's the link to the tweet

https://nitter.net/AnnaIndianaAI/status/1728089499429642432#m

Yeah the music is trash but so are the songs these days. The AI perfectly learned the pleasures of conformity and the importance of trends.

I think in the context of K-Pop it makes total sense, the music and everything around is anyway just done after a formula which has proven to work very well to sell. While right now you need to put children and teenagers through years of rigorous training and expose them to immense stress and pressure so most of them break, with AI you can easily replicate the same formula and refine much quicker without throwing so many young people into the meat grinder of the music industry.

More money and control for the companies less people killing themselves.

The ones who really burn for the music will make music despite AI music being available. And they also will find an audience, even though it might be smaller.

Watching the video and the head movement all I see is the Balenciaga ai meme

YouTube link for those that want to see it for themselves. ~~https://m.youtube.com/watch?v=3T175LTQFnw~~ https://www.youtube.com/watch?v=0l71Hh1gPkE

To me the biggest problem with it is that it doesn't understand the relationship between the meanings of the words and the melody of the song. It kinda makes it sound like a bad parody song. I think if you looked at just the lyrics or just the melody they would be quite convincing on their own.

I think it sounds the same as other shitty music I hear all the time.

I fully expect AI music to be part of the musical scene going forward.

Does it feel dystopian? Yeah, it's not a fun future. But this is here to stay.

Just looking at the picture this already seems like amateur work.

As for the music being ... Yes it's pretty flat and unexciting. Realistically lots of today's music is already a remix of a remix and honestly... quite generic. Bands have been cast for looks for a long time, autotune has been here for ages, and with a better effort in the voice department I don't see AI artists being far off. I don't think they will ever completely replace artists here, especially since live events really are a thing of their own but.. eh

Here I am just wondering how much automation went into tuning/tweaking the voice. I've made some stuff in Synthesizer V Studio and that's a big chunk of the work. Being able to load something into an AI and have it give a pass would be awesome.

I don't understand why people are so cynical about this, it seems like a harmless demonstration of the current state of the technology.

Waiting for the AI composer site where I can prompt it for music based on a story I provide.