omg your art is so am(A)z(I)ng!!!

 grok, make my math homework naked and cover it in yogurt

ok look. before i start this post, i’m going to make the assumption that no one, including you and me, likes ai generated art, videos, images, whatever the fuck genAI shits out. we’re woke. we know the environmental impacts of ai, we know how many jobs it’s killing, we know every negative fact about ai there is to know because we hate it. as we should. hunk of junk rustbucket clanker. cool, we got that out of the way.

then why the fuck do we like ai generated music so much?

dead ass. you’ve been listening to ai music, whether you know it or not. not only that, but i hear so many arguments within the vocaloid community defending ai music. i’m talking about the red one. yknow like the really fat one

i think teto is ai music in some capacity. 

DON’T SHOOT HOLD YOUR FIRE

i know it already sounds like an L take. but please just hear me out. it’s well known that synthV teto uses ai to help producers. the ai makes teto sound more realistic. it connects words and smooths out transitions. clearly, synthv is an ai tool, right? why does the vocaloid community not care at all that their favorite songs were made with ai? 

well, there’s a lot of positives with synthVs and other voice softwares with ai enhancement, such as kafu from cevio. 

1. consent. a lot of the arguments online state that synthV is ethical because it uses samples from consenting vocalists. honestly, great argument.

2. environmental impact. unlike chatgpt and other LLMs, this ai is stored entirely on your computer. not on the cloud, not in some water guzzling datacenter, it’s on your hard drive. as such, the only workload is being handled by your cpu, which has a very negligible impact on the environment. also a good argument.

the issue i have with defenders is that no one is claiming that their music is ai, even though it is. here’s an example that i often like to use: imagine an artist is making a piece for an art contest. they’ve done 70% of the work. they ask chatgpt to do the other 30%. is this ai art? what if chatgpt did 20%? 10%? 1%? and if you don’t like chatgpt, what about any other art software with built in ai features? what if the ai model was made ethically and with consent? is it still fair for that person to win the art contest? if you answered no, is it fair for overdrive to win best vocaloid song at music awards japan? (true story: while it didn’t win, it was a nominee along with tetoris.)

cuz here’s my issue with it. while there are no moral concerns with this software, there are artistic and creative concerns. why should an artist who uses ai for their art compete with another artist who used 0 ai? take mitchie m, one of the godliest tuners of all time. no doubt hours poured into making miku, this old, robotic voicebank, sound realistic. we should acknowledge that effort and how much more skill it takes to tune miku than synthv teto. here’s my request: on the product page for project diva: megamix on steam, there is a disclaimer at the bottom that the game contains ai generated content because the dlc has teto. i request that all past, current, and future synthv teto songs should have an ai disclaimer in the description or title if published on youtube, spotify, etc. if you’re gonna use teto, acknowledge how she works. it’s only fair to the people listening as well.

i’m obviously not saying stop supporting and stop listening to teto songs. i love teto songs! go listen to cheap properties! listen to bad taste! listen to machine love! listen to phony! listen to miku v6! don’t listen to the distasteful addition of ai tupac in drake’s diss track! i love art just like any other normal person. 

but the next time you see something generated using an ai and immediately call it slop, pause and think. what is it that you actually hate about it?

- is it that it looks bad? like it or not, as we keep feeding ai more and more info, ai is just gonna keep getting better to a level not necessarily above humans but eventually comparable. 

- is it that it was generated without consent of the artists it steals? yeah, we need guidelines and regulations and lots and lots of laws. but if consent was acquired, would you still call it slop?

from what i’ve read (which are admittedly just dozens of reddit posts and youtube comments), the art community is divided and, in my opinion, somewhat uneducated. some say that gen ai needs to be regulated. some say it needs to be banned. gen ai is a broad term, and as ai evolves, so too does that definition. synthv is generating teto’s voice based off samples, so it could be classified as gen ai.

here’s my second call to action. the art community needs to come to a consensus about what it wants and what it is asking for. what reasons in particular does ai need to be regulated? obviously, 100% human made art is good and 100% ai made art sucks. but what about the gray area that is synthv? we need to stop mindlessly calling everything slop; it’s just too lazy and easy, and we’re better than that. i want to be on your side, but i want a clear argument presented so we can all agree and stand up to ai. 

i apologize for writing about this. in my first blog, i said i wouldn’t talk about topics i am not specialized in, and i am not a visual artist. however, i am ai certified by nvidia in ai infrastructure and operations, if that gives me any credibility. i felt deeply compelled to write this because i don’t see anyone else addressing or stating this take. i would love to know your thoughts or counterarguments. and hell, i would love if you shared this with other people if you feel compelled to.

and btw; this is not a defense for ai “artists”. prompting is not art, you’re vibe-drawing a picture. go goon to your ai girlfriend