New Igorrr! A pleasant surprise. I’m torn on the AI usage in the video. I’ve been very vocally opposed to AI art in previous posts in my time on Tildes, but for once I feel like this is starting...
New Igorrr! A pleasant surprise.
I’m torn on the AI usage in the video. I’ve been very vocally opposed to AI art in previous posts in my time on Tildes, but for once I feel like this is starting to skirt the line towards usage that I think is acceptable, because it truly felt like a tool used by an expert in the field vs. a crutch used by someone with no inherent CGI ability. In other words, I don’t think an amateur could re-create this just by vibing and prompting - I think you need a real technical ability to be able to use AI as a foundation like this and build on it.
Reason I say all this is because this same producer has made zero-AI CGI videos for this same musician in the past, so I know they’re perfectly capable of the medium sans AI, and I didn’t notice any super obvious (extra fingers, visual “tearing,” etc) AI artifacts, which leads me to believe that they did some post-processing after the fact to clean it up, again showcasing their abilities to still polish and deliver a final product.
As far as the music goes, sounds like typical Igorrr to me (meant in a good way) and I’m looking forward to seeing what else he puts out soon.
What are beginners allowed to do? No snark, serious question. In music, for example, there are beginner's instruments, which ideally are of decent quality, but without the expense and small...
I think is acceptable, because it truly felt like a tool used by an expert in the field vs. a crutch used by someone with no inherent CGI ability.
What are beginners allowed to do? No snark, serious question. In music, for example, there are beginner's instruments, which ideally are of decent quality, but without the expense and small details that only matter to more experienced players.
No snark taken, and I appreciate the discourse! I might be misunderstanding your question, because I don't really care about the inherent quality of the tool being used, so long as, if it's being...
No snark taken, and I appreciate the discourse! I might be misunderstanding your question, because I don't really care about the inherent quality of the tool being used, so long as, if it's being marketed to folks as a finished product (more on that later), it's still enjoyable to consume as an end user/consumer. There are plenty of self-recorded bangers in the world played on mediocre instruments, but the music itself is so good that it doesn't matter.
I'll instead answer the separate question of "how are beginners supposed to learn if Shevanel is suggesting they're not allowed to use the tools in the first place?" since you used the word "allowed" - again, apologies if I'm not representing your question correctly here; it's late in the US where I'm at so I may have to take another swing at this tomorrow if I'm totally missing the mark.
My answer to that separate question is that I believe that "beginners" in this kind of space should absolutely, positively, create as much as they possibly can. Go wild with it, beat your API key into submission, make your mistakes, make your progress, and have fun with the whole process... and at the end of these first steps into this new creative process, do NOT, for the love of all that is holy, immediately share your product. At least, not for now.
My primary complaint, as I've outlined on Tildes before, is not that this medium exists. It's that folks think that because "now I can make an art too," that it behooves them to share it out with the world, and we are absolutely flooded with zero-thought, mediocre AI-generated content, instead of gems that showcase AI actually being used to its well-honed potential by artists who were already excelling in the space before the introduction of AI.
You brought up beginning instrument players, so I'll make a comparison back to that - I'm a musician myself, and after spending a lot of time honing my craft, I am comfortable and willing to record myself (or be recorded) and have that material exist out in the world to be consumed by other enjoyers of music. That said, I would be incredibly frustrated if suddenly, Spotify started filling up with recordings of 5th grade band recitals. But that's exactly what's happening with AI art - people learn one iota of the topic and then just start blasting. It's important that these recitals happen, it's important that we keep supporting them and that kids keep learning music in this way, but it's disingenuous to want to try to publish and market it as if it's a finished product that ought to be weighed in a medium such as an audio streaming service. The same goes for beginners in the gen AI space as well.
All of that said, I again want to acknowledge that the learning process is vital to any sort of art form, but for goodness's sake, you don't need to share every single piece of it with the world at large every time you do something. In that same vein, I think anyone utilizing gen AI (myself included) is doing the medium a disservice by not making it a point to gather in more collective educational spaces (think forums/subreddits) and share out / mentor each other in these places where folks can grow and learn together. Because yeah, if that space existed, folks should absolutely feel comfortable sharing their works in progress to their heart's content, because that's the purpose of it. And you'd be totally mad to barge into that space and complain about the mediocrity within, just like you'd be mad to barge into a 5th grade band room and say that the kids suck at their instruments.
Maybe the above spaces do exist and they’re just not making much of a name for themselves; please excuse my ignorance if so. My point is that folks ought to try to constrain the “work in progress” to more spaces like that.
Two other thoughts that don't really fit anywhere else:
I think that artists who already have made a name for themselves and/or who are more familiar with the medium in general are, by the very nature of their prior experience, going to get more of a pass because they'll use their knowledge transfer to get ahead of the raw beginners in the same space. It's the same reason why watching someone like Rob Scallon learn an instrument is still rewarding and fun, because he builds on his prior musical skillset and applies it to a new instrument, whereas I'd rather get a root canal than listen to a 13-minute video of a 5th grader learning cello scales (and I say this as a former music teacher). In the same vein, this is just like how a well-established CGI artist is going to grep AI video generation more easily.
There's no nice way to say this, so I'll just be blunt about it - if folks new to the gen AI space want to keep polluting "curated" mediums (quotes getting ever bigger every year) with their slop, they need to be ready to face serious scrutiny upon doing so. Back to my music analogy, I'm a drummer. I like to think I'm a pretty good drummer. I am not a good violinist. If I showed up to an open jam with a violin, and tried to play it alongside a bunch of other legitimate musicians, I would expect to be razzed about it. Hell, I might be asked to leave the jam and not come back. Play stupid games, win stupid prizes. AI art is interesting, and people taking advantage of it are increasingly doing cool things with it, but it doesn't just get a pass because of what it is. I'm going to judge it just like I'm going to judge the art form before AI was implemented within it.
I hope all of this makes sense! I really do think the medium is promising, but I think society needs to make a distinction between "I am capable of creating art now" and "All of the things I create are valuable and ought to be shared," especially when the creation of said art is near-instantaneous now and the flooding of the medium is so much more easy at this point vs. even a few years ago.
Though I’m not entirely on the same page (I feel that use of AI in the arts in many ways is questionable, but those in power continue to choose to ignore the questions), this is a well written...
Though I’m not entirely on the same page (I feel that use of AI in the arts in many ways is questionable, but those in power continue to choose to ignore the questions), this is a well written reply.
I have a hunch, though, that if the users currently posting slop were required to stop until they became more capable and thoughtful that they’d simply stop using AI altogether. I believe that for many, the point of attraction is not the ability to create as much as it is AI’s role in a karma/like pipeline. They don’t care about what they create as long as it allows them to push a button and receive the attention they crave.
This would line up with the contempt I’ve seen many AI proponents holding against artists online. These people see artists posting their work and receiving positive attention and feel cheated because they don’t see all the experience and work that goes into creating those pieces and don’t understand why the artists are receiving attention instead of themselves. Even prior to the existence of AI, they felt like artists had their own push-button karma machine that nobody else can use, even though that’s not at all the truth.
Appreciate your thoughts! I totally hear you, and deep down I feel much the same way as you do. The entire medium is built on an impossibly large mountain of intellectual property theft, and by...
Appreciate your thoughts! I totally hear you, and deep down I feel much the same way as you do. The entire medium is built on an impossibly large mountain of intellectual property theft, and by its very nature the art in and of itself (without outside intervention after its creation) is going to remain reductive and shallow. Yet at this point, I’ve accepted that we’re not going to get rid of it, so I’ve moved on to, “okay, so how do we acknowledge and contain it so it doesn’t just drown out everything else?”
Your last paragraph hits on a great point. It’s akin to that idea of people being shocked that a live band wants to charge $500/hr to play at their event. You’re not paying them for the x hours that they’re at the event. You’re paying them for the tens of thousands of hours that went into their craft for them to be able to show up to an event and play music at that level. The same goes for visual artists, of course; I just lack the familiarity with that skillset to be able to speak to it in the same way.
Also:
if the users posting slop were required to stop until they became more capable and thoughtful, […] they’d simply stop using AI altogether
Personally I don't think GenAI tools will ever dig themselves out of the hole they're in. I think it's true that in the right hands they could probably be used to create stuff of actual value, but...
Personally I don't think GenAI tools will ever dig themselves out of the hole they're in. I think it's true that in the right hands they could probably be used to create stuff of actual value, but that's not what the tools themselves are aspiring to be. You look at any marketing spiel for these things (remember the awful Apple Intelligence ads?) and they're all but shouting from the rooftops "do awesome things with no effort even if you're a schlub with no skill," so that's the kind of user base they want to and will attract, fueling the reputation that GenAI produces nothing but worthless slop, and that whole negative feedback loop will scare serious players (both tool makers and artists) from exploring the possibility that they can be used for anything else.
edit: I guess I should add that while I feel that way about GenAI tools, I also expect that the general population's expectations may eventually lower so much that it doesn't matter and we'll come to accept the slop as the new normal. So I guess I'm a cynic at both ends of the equation.
I hear you, and I think the cynicism is warranted. At this point I almost view it like programmed drums in metal music. This is a very niche tangential complaint so it won’t land with most, but I...
I hear you, and I think the cynicism is warranted.
At this point I almost view it like programmed drums in metal music. This is a very niche tangential complaint so it won’t land with most, but I can almost always tell when drums are programmed in an album (vs. recorded live), and it happens a lot more in metal music, likely due to the general dynamic stagnation and consistency of the parts vs other genres. I don’t like when groups do it, but I acknowledge that for many, the cost, talent, and time factors are insurmountable, so they turned towards canned drums as a solution. That said, when these groups take their act on tour, I sincerely hope they pony up the cash to hire a drummer that can play their parts. And if they wrote impossible parts and/or refuse to shell out for a drummer who can hang, they have to sleep in the shitty bed they made. Same goes for groups who simply write terrible drum parts because they don’t get any feedback from their auto-drummer. They need to be prepared to be criticized for their poor use of a musical shortcut.
I hope groups that use canned drums do record with live drums in the future, as they find the time or funds, or simply come to the realization that the active collaboration far outweighs the perceived benefits of a canned drummer.
In the same vein(s), I’m going to be disappointed if, for example, the video producer that Igorrr uses continues to only use an AI palette moving forward, and I think it would lead to creative bankruptcy pretty quickly. But if it’s a single item in a larger toolbelt? It’s a pill I can swallow every so often if used in the hands of folks who know what they’re doing. We’re never going to wish this medium away, but I hope we can strictly relegate it to the category of “fortifying pre-existing art” vs. “Acting like it can replace artists whole-cloth.”
Reminds me of Aphex Twin and Squarepusher a bit. I know glitchy, experimental music like this is not everyone's cup of tea, but I love it. It makes me feel kinda weird listening to it, like I'm on...
Reminds me of Aphex Twin and Squarepusher a bit. I know glitchy, experimental music like this is not everyone's cup of tea, but I love it. It makes me feel kinda weird listening to it, like I'm on something. :P The bizarre AI generated music video actually suits it really well too, IMO.
Igorrr! Love him. I'm indifferent to the video portion of this, but the music is great, makes me think the new album will be quite promising. 'Savage Sinusoid' is one of my all-time favorite...
Igorrr! Love him. I'm indifferent to the video portion of this, but the music is great, makes me think the new album will be quite promising. 'Savage Sinusoid' is one of my all-time favorite albums, but I felt 'Spirituality and Distortion' was a weaker effort. I hope this new one leans into the Squarepusher-y breakbeat type of stuff that dominates the first half of this song.
I’ve enjoyed all his releases, but I’ll always have a soft spot in my heart for “Hallelujah,” primarily because it was my introduction to Igorrr. How does that one stand up to the others in your mind?
I’ve enjoyed all his releases, but I’ll always have a soft spot in my heart for “Hallelujah,” primarily because it was my introduction to Igorrr. How does that one stand up to the others in your mind?
Better than S&D, but not quite as good as 'Sinusoid'. I suspect his first couple of releases were a process of trying to figure out how to fit everything together - which is hard when 'everything'...
Better than S&D, but not quite as good as 'Sinusoid'. I suspect his first couple of releases were a process of trying to figure out how to fit everything together - which is hard when 'everything' consists of black metal, breakcore, classical music, French folk music, up to and including the proverbial kitchen sink. 'Hallelujah' has a few really great tracks but a lot of it still sounds slightly arbitrary, like it's asking 'what could I put here' as opposed to 'what should I put here'.
Sinusoid, on the other hand, feels more in touch with the structural or even 'music theoretical' aspects of each of its parts, so that even though on the surface they may seem arbitrary (chiptune synths over blast beats?) there is a deeper connection that creates a surprisingly consistent aesthetic. Actually, I feel like S&D does about as well as Sinusoid in this respect, but at the same time it was more restrained or even formulaic with its approach. And fundamentally, I'd rather an album take big swings (even if they're not all winners) than play it safe - though of course with Igorrr, 'safe' is a relative word.
Haha that's fair; I think the kitchen-sink approach is what I liked so much about "Hallelujah," though its cohesion definitely does take a hit because of it. I've dug your recommendations in the...
Haha that's fair; I think the kitchen-sink approach is what I liked so much about "Hallelujah," though its cohesion definitely does take a hit because of it.
I've dug your recommendations in the past, so I'll pose another question if you'll humor me - do you know of any other acts that scratch the same itch that Igorrr does? It's a very specific itch, granted :) you did mention Squarepusher and I've really only checked out one of their albums; they're probably worth a revisit at the very least.
rubs hands gleefully Okay, let's see... It's hard to find anything exactly the same as Igorrr due to how unique his approach is. But what you've labeled an 'Igorrr itch' might more generally be...
rubs hands gleefully
Okay, let's see... It's hard to find anything exactly the same as Igorrr due to how unique his approach is. But what you've labeled an 'Igorrr itch' might more generally be for music that 1) blends lots of styles, focusing especially on electronic/breakbeat and metal, 2) is really fast and complicated, with so many ideas rushing at you so quickly that it's hard to keep up, though if you manage to, the feeling is exhilarating, and 3) has a sense of humor, plays 'musical jokes' on you.
Items 2) and 3) are actually more related than you might think. The essence of comedy is the unexpected, and the more syncopated and faster music is, the harder it is to expect the next note. Yet well-written music may also have the overarching logical structure necessary for humor (what in jokes often boils down to double meanings). I often find myself laughing at fast, clever music - I remember hearing Archspire's 'Relentless Mutation' for the first time with some friends, and we were all cracking up (that might actually be a good recommendation, come to think of it, though it's more purely metal than Igorrr). I also remember a few people laughing at a show for Squarepusher's live band, Shobaleader One, for more or less the same reasons, I imagine.
And indeed, Squarepusher might scratch this itch pretty well, but he has a big, diverse discography, so selection is important. His early releases may have the requisite speed and complexity, but by today's standards they're kind of conventional, and the timbral palette is probably not diverse enough. I think the first really 'way out there' Squarepusher album, and my personal favorite, is 'Ultravisitor'. Certain tracks might scratch the itch, but there might be too many slow, almost ambient ones in between. So maybe a better suggestion is 'Ufabulum', which is more uniformly high-intensity, but maintains a really sophisticated and experimental timbral approach. It's purely electronic, though I feel like certain tracks have a metal-ish vibe to them. 'Damogen Furies' might also be a good pick - quite similar to 'Ufabulum', perhaps even more intense, though also less nuanced.
Okay, on to other recommendations - check out 'Power Tools' by Big Lad. It's a fusion of grindy powerviolence style hardcore and breakcore or gabber electronic elements. Super fun, and has a way of, just when it feels like its intensity is peaking, getting more intense. No vocals though. A similar pick to this might be drumcorps, who has a bunch of great EPs, my favorite being 'Better Days' - a mathcore/metalcore/breakcore fusion with digitally fucked-with harsh vocals - lots of over-the-top fun. And then there is Machine Girl, who uses a lot more synths than these other two, and whose songwriting is a little more straightforward, but still brings a lot of metal tropes and is a great, somewhat irreverent vocalist imo. 'U-Void Synthesizer' is my favorite of theirs.
There are also quite a few electronic artists who don't have much of a metal influence, but still have intense, fast, and inventive songwriting, along with stylistic diversity, which I think could fit the bill. Iglooghost's latest releases have been a bit milder, but their breakout album 'Neo Wax Bloom' is a great piece of future bass/IDM with tons of energy and overflowing creativity. Lynyn is the solo project of the guitarist from the band Monobody, and his album 'lexicon' is really good, very textural in a way that's reminiscent of Squarepusher but also more modern. A left-field pick might be Gazelle Twin, whose album 'Pastoral' is really incredible - a mix of British folk music, creepy dark electronica, and aggressive noise. It's definitely way less intense than Igorrr on the whole, but might make up for it with sheer weirdness and, at times, dark humor.
Oh, and I almost forgot MIRAR's 'Ascension' - this just came out and I'm still digesting it, but it's this very unusual mixture of sludge metal riffs, dubstep's screechy noises and paced out rhythms, a little drum 'n' bass, flavor and then an out-of-nowhere dose of classical music. I think maybe on paper this might be the closest to Igorrr, but I'm not quite sure it meets the mark quality-wise, at least compared to my personal opinions on the rest of these releases.
Really, nothing's going to compare exactly to Igorrr, but hopefully you find something you enjoy here. Let me know what you think! If you like something in particular I might be able to provide more targeted recommendations.
Love this; thanks as always for sharing! You brought up a great point about the musical surprises making one laugh - I actually think I’ve had the exact same experience with Archspire myself, I...
Love this; thanks as always for sharing!
You brought up a great point about the musical surprises making one laugh - I actually think I’ve had the exact same experience with Archspire myself, I totally get it!
Outside of that act, I’m also familiar with Mirar’s work too. On the whole, I preferred their Mare EP to Ascension, though I still enjoyed Ascension and do need to spend more time with it. Mirar’s work is all “thall-adjacent” and is heavily influenced by groups like Vildhjarta and Humanity’s Last Breath, two other groups that I really like that definitely have a lot of the same aggressive approach that can veer into a different direction unexpectedly and make you laugh. Big fan!
On the more electronic side, I enjoy Venetian Snares for much the same reason - breakcore is already chaotic, and he takes it to that next level while still providing musical “through-lines” that keep his pieces from ever descending into pure noise.
Outside of those couple of groups, virtually everything else you’ve mentioned is new to me, so I’ve got some work to do. I’ll poke around and be sure to keep you posted; thanks again!
I was going to respond to the @eyechoirs comment about including ol' Venetian of the Snares, but didn't think it quite fit in their fantastic suggestions. Happy you brought them up! Venetian...
On the more electronic side, I enjoy Venetian Snares for much the same reason - breakcore is already chaotic, and he takes it to that next level while still providing musical “through-lines” that keep his pieces from ever descending into pure noise.
I was going to respond to the @eyechoirs comment about including ol' Venetian of the Snares, but didn't think it quite fit in their fantastic suggestions. Happy you brought them up! Venetian Snares is up there as one of my favorite artists, period.
VSnares is great! I probably should have mentioned Rossz Csillag Alatt Született (which might be my favorite album of his) due to the heavy use of classical music samples, which I guess is...
VSnares is great! I probably should have mentioned Rossz Csillag Alatt Született (which might be my favorite album of his) due to the heavy use of classical music samples, which I guess is adjacent to Igorrr.
I saw this link, then watched the video and saw "Meat Dept." I was immediately stoked, I saw their VERY NOISE video which led me to their short film Black Holes, which is a pretty great animation....
I saw this link, then watched the video and saw "Meat Dept." I was immediately stoked, I saw their VERY NOISE video which led me to their short film Black Holes, which is a pretty great animation.
This has their style all over it, assisted by whatever they did with genAI. From their description there seems to be initial 3D renders enhanced by genAI, which is also sort of shown by the design they use (there's sort of an exaggeration you can see in Black Holes that's present here). I thought it was an interesting video, and thought the sort of soulless AI vibe helped with the content a bit.
Not specifically, Igorrr makes Metal-Focused IDM/Breakcore ("Intelligent Dance Music", a tongue in cheek response to "EDM"). The music is certainly made to be on its own, they just also make some...
Not specifically, Igorrr makes Metal-Focused IDM/Breakcore ("Intelligent Dance Music", a tongue in cheek response to "EDM"). The music is certainly made to be on its own, they just also make some really fantastic videos (this one uses AI + Traditional Rendering, whereas the others use just traditional rendering).
It's certainly a type of genre that isn't for everyone, personally it's one of my favorites, but its chaotic and very different than anything else here. It's the breakbeat, drum focused, glitched version of freeform jazz, essentially.
Still has that uncanny-valley effect and visible artifacting, but I suppose it works better for their style; the last video of theirs that I've seen, VERY NOISE, was very, uh, interesting.
Still has that uncanny-valley effect and visible artifacting, but I suppose it works better for their style; the last video of theirs that I've seen, VERY NOISE, was very, uh, interesting.
With the title of "ADHD", as someone with severe ADHD, I was curious. I'm still confused as to what exactly the message is supposed to be. Although with projects like this, I know often they will...
With the title of "ADHD", as someone with severe ADHD, I was curious.
I'm still confused as to what exactly the message is supposed to be. Although with projects like this, I know often they will leave it up in the air a bit. I could definitely see some possible "ADHD is overdiagnosed and medications turn people crazy", but also perhaps just some representations of what ADHD can feel like to have. i.e., a negative and positive approach to the messaging.
I don't know. I have mixed feelings about what's being portrayed, but my instinct is that it is not positive, which is annoying. I deal with enough negativity in my life thanks to ADHD and to people not understanding ADHD that it's always annoying to see this sort of thing. Like how, as a geek, seeing the decades of movies and shows with the trope of nerdy geeks being socially awkward and weird doesn't reflect the reality that - yes, we can sometimes or even often be socially awkward and/or weird, but most people can't identify geeks by sight like you can in a movie. That trope gets… tiring. In the same way, ADHD tropes are also tiring.
I don't require the conversation about ADHD to be positive. It is, in fact, a neurological disability, and while I'm not totally unhappy having it; while there are positives and things about it that help make me who I am; at the same time, I have paid so much ADHD tax in my life, and I will probably die sooner than I would otherwise thanks to ADHD. That being said, I do appreciate the positive conversations about ADHD - how to understand those of us that have it; how we can understand the neurotypical folks; learning coping techniques to minimize the negative impacts.
Meanwhile, on the music from this artist… hadn't heard of them before, not sure I'll go out of my way to find them. I hardly considered this "music" — that said, that's personal preference I'm expressing. I can recognize and appreciate the technical expertise here. They are good at composition, and if this style is up your alley, this seems to be a good example of it. So while it's not for me, it seemed pretty good otherwise.
The AI was an interesting use. I think what interested me most was the ability to have recurring characters. And the technical skill to be able to get something that they wanted rendered. I can imagine this took a LOT of work. Pulling numbers completely from my ass, I would imagine that they surely had to generate a lot of clips to get ones that had what they wanted. It's rather impressive. And from a visual standpoint - the music of the eyes - it was visually appealing. I really liked the colour grading, for example. It really had that 50s era feel. I'm not against seeing more AI. It's useful already for surreal projects like this. I can't imagine what another 10-15-20 years will bring us.
It reminds me of how people have been against recorded media - records, movies - because you can't beat live performances. But movies and albums have been wildly successful. I think we will enter an era where we still have movies; still have live theatre; but also have AI-generated media that will allow for storytelling that is truly amazing. To allow my ADHD mind to follow yet another rabbit hole: People claim to hate CGI. It is my humble opinion that what most people hate is bad CGI. Good CGI? You don't notice it. Like good practical effects. So consider once we get to the point of AI being able to generate stuff that is longer, with dialogue, to the point where you can't tell it apart from live actors… we'll be able to have shows without the "alien of the week" problem, i.e. aliens who have to be in relatively human-sized costumes because humans play them. We can have truly alien beings interacting with humans because it's all generated.
I think there will be room for that, along with more traditional movies with CGI and things like live theatre, if that makes sense.
Gautier Sierre (Igorrr) has a note on the video about the song Essentially, the goal is to express disjoint, loosely connected passages similar to how one might have issues focusing on a single...
Gautier Sierre (Igorrr) has a note on the video about the song
Kind of an autobiographical piece of music. Starting from one point and moving to another, with no clear link except for the person itself. From simple thoughts, symbolized here as simple dots of sound in the silence, to a complex pathological chaos that somehow still stands. It’s getting worse and worse until the final giant lets go.
Essentially, the goal is to express disjoint, loosely connected passages similar to how one might have issues focusing on a single thing with ADHD. From his description it's autobiographical, which may help contextualize it a bit for you.
New Igorrr! A pleasant surprise.
I’m torn on the AI usage in the video. I’ve been very vocally opposed to AI art in previous posts in my time on Tildes, but for once I feel like this is starting to skirt the line towards usage that I think is acceptable, because it truly felt like a tool used by an expert in the field vs. a crutch used by someone with no inherent CGI ability. In other words, I don’t think an amateur could re-create this just by vibing and prompting - I think you need a real technical ability to be able to use AI as a foundation like this and build on it.
Reason I say all this is because this same producer has made zero-AI CGI videos for this same musician in the past, so I know they’re perfectly capable of the medium sans AI, and I didn’t notice any super obvious (extra fingers, visual “tearing,” etc) AI artifacts, which leads me to believe that they did some post-processing after the fact to clean it up, again showcasing their abilities to still polish and deliver a final product.
As far as the music goes, sounds like typical Igorrr to me (meant in a good way) and I’m looking forward to seeing what else he puts out soon.
What are beginners allowed to do? No snark, serious question. In music, for example, there are beginner's instruments, which ideally are of decent quality, but without the expense and small details that only matter to more experienced players.
No snark taken, and I appreciate the discourse! I might be misunderstanding your question, because I don't really care about the inherent quality of the tool being used, so long as, if it's being marketed to folks as a finished product (more on that later), it's still enjoyable to consume as an end user/consumer. There are plenty of self-recorded bangers in the world played on mediocre instruments, but the music itself is so good that it doesn't matter.
I'll instead answer the separate question of "how are beginners supposed to learn if Shevanel is suggesting they're not allowed to use the tools in the first place?" since you used the word "allowed" - again, apologies if I'm not representing your question correctly here; it's late in the US where I'm at so I may have to take another swing at this tomorrow if I'm totally missing the mark.
My answer to that separate question is that I believe that "beginners" in this kind of space should absolutely, positively, create as much as they possibly can. Go wild with it, beat your API key into submission, make your mistakes, make your progress, and have fun with the whole process... and at the end of these first steps into this new creative process, do NOT, for the love of all that is holy, immediately share your product. At least, not for now.
My primary complaint, as I've outlined on Tildes before, is not that this medium exists. It's that folks think that because "now I can make an art too," that it behooves them to share it out with the world, and we are absolutely flooded with zero-thought, mediocre AI-generated content, instead of gems that showcase AI actually being used to its well-honed potential by artists who were already excelling in the space before the introduction of AI.
You brought up beginning instrument players, so I'll make a comparison back to that - I'm a musician myself, and after spending a lot of time honing my craft, I am comfortable and willing to record myself (or be recorded) and have that material exist out in the world to be consumed by other enjoyers of music. That said, I would be incredibly frustrated if suddenly, Spotify started filling up with recordings of 5th grade band recitals. But that's exactly what's happening with AI art - people learn one iota of the topic and then just start blasting. It's important that these recitals happen, it's important that we keep supporting them and that kids keep learning music in this way, but it's disingenuous to want to try to publish and market it as if it's a finished product that ought to be weighed in a medium such as an audio streaming service. The same goes for beginners in the gen AI space as well.
All of that said, I again want to acknowledge that the learning process is vital to any sort of art form, but for goodness's sake, you don't need to share every single piece of it with the world at large every time you do something. In that same vein, I think anyone utilizing gen AI (myself included) is doing the medium a disservice by not making it a point to gather in more collective educational spaces (think forums/subreddits) and share out / mentor each other in these places where folks can grow and learn together. Because yeah, if that space existed, folks should absolutely feel comfortable sharing their works in progress to their heart's content, because that's the purpose of it. And you'd be totally mad to barge into that space and complain about the mediocrity within, just like you'd be mad to barge into a 5th grade band room and say that the kids suck at their instruments.
Maybe the above spaces do exist and they’re just not making much of a name for themselves; please excuse my ignorance if so. My point is that folks ought to try to constrain the “work in progress” to more spaces like that.
Two other thoughts that don't really fit anywhere else:
I think that artists who already have made a name for themselves and/or who are more familiar with the medium in general are, by the very nature of their prior experience, going to get more of a pass because they'll use their knowledge transfer to get ahead of the raw beginners in the same space. It's the same reason why watching someone like Rob Scallon learn an instrument is still rewarding and fun, because he builds on his prior musical skillset and applies it to a new instrument, whereas I'd rather get a root canal than listen to a 13-minute video of a 5th grader learning cello scales (and I say this as a former music teacher). In the same vein, this is just like how a well-established CGI artist is going to grep AI video generation more easily.
There's no nice way to say this, so I'll just be blunt about it - if folks new to the gen AI space want to keep polluting "curated" mediums (quotes getting ever bigger every year) with their slop, they need to be ready to face serious scrutiny upon doing so. Back to my music analogy, I'm a drummer. I like to think I'm a pretty good drummer. I am not a good violinist. If I showed up to an open jam with a violin, and tried to play it alongside a bunch of other legitimate musicians, I would expect to be razzed about it. Hell, I might be asked to leave the jam and not come back. Play stupid games, win stupid prizes. AI art is interesting, and people taking advantage of it are increasingly doing cool things with it, but it doesn't just get a pass because of what it is. I'm going to judge it just like I'm going to judge the art form before AI was implemented within it.
I hope all of this makes sense! I really do think the medium is promising, but I think society needs to make a distinction between "I am capable of creating art now" and "All of the things I create are valuable and ought to be shared," especially when the creation of said art is near-instantaneous now and the flooding of the medium is so much more easy at this point vs. even a few years ago.
Thank you for a fantastic response! I'll add more tomorrow when I'm not so tired, but what a great writeup. Thanks again.
Though I’m not entirely on the same page (I feel that use of AI in the arts in many ways is questionable, but those in power continue to choose to ignore the questions), this is a well written reply.
I have a hunch, though, that if the users currently posting slop were required to stop until they became more capable and thoughtful that they’d simply stop using AI altogether. I believe that for many, the point of attraction is not the ability to create as much as it is AI’s role in a karma/like pipeline. They don’t care about what they create as long as it allows them to push a button and receive the attention they crave.
This would line up with the contempt I’ve seen many AI proponents holding against artists online. These people see artists posting their work and receiving positive attention and feel cheated because they don’t see all the experience and work that goes into creating those pieces and don’t understand why the artists are receiving attention instead of themselves. Even prior to the existence of AI, they felt like artists had their own push-button karma machine that nobody else can use, even though that’s not at all the truth.
Appreciate your thoughts! I totally hear you, and deep down I feel much the same way as you do. The entire medium is built on an impossibly large mountain of intellectual property theft, and by its very nature the art in and of itself (without outside intervention after its creation) is going to remain reductive and shallow. Yet at this point, I’ve accepted that we’re not going to get rid of it, so I’ve moved on to, “okay, so how do we acknowledge and contain it so it doesn’t just drown out everything else?”
Your last paragraph hits on a great point. It’s akin to that idea of people being shocked that a live band wants to charge $500/hr to play at their event. You’re not paying them for the x hours that they’re at the event. You’re paying them for the tens of thousands of hours that went into their craft for them to be able to show up to an event and play music at that level. The same goes for visual artists, of course; I just lack the familiarity with that skillset to be able to speak to it in the same way.
Also:
You say this like it’s a bad thing ;)
Personally I don't think GenAI tools will ever dig themselves out of the hole they're in. I think it's true that in the right hands they could probably be used to create stuff of actual value, but that's not what the tools themselves are aspiring to be. You look at any marketing spiel for these things (remember the awful Apple Intelligence ads?) and they're all but shouting from the rooftops "do awesome things with no effort even if you're a schlub with no skill," so that's the kind of user base they want to and will attract, fueling the reputation that GenAI produces nothing but worthless slop, and that whole negative feedback loop will scare serious players (both tool makers and artists) from exploring the possibility that they can be used for anything else.
edit: I guess I should add that while I feel that way about GenAI tools, I also expect that the general population's expectations may eventually lower so much that it doesn't matter and we'll come to accept the slop as the new normal. So I guess I'm a cynic at both ends of the equation.
I hear you, and I think the cynicism is warranted.
At this point I almost view it like programmed drums in metal music. This is a very niche tangential complaint so it won’t land with most, but I can almost always tell when drums are programmed in an album (vs. recorded live), and it happens a lot more in metal music, likely due to the general dynamic stagnation and consistency of the parts vs other genres. I don’t like when groups do it, but I acknowledge that for many, the cost, talent, and time factors are insurmountable, so they turned towards canned drums as a solution. That said, when these groups take their act on tour, I sincerely hope they pony up the cash to hire a drummer that can play their parts. And if they wrote impossible parts and/or refuse to shell out for a drummer who can hang, they have to sleep in the shitty bed they made. Same goes for groups who simply write terrible drum parts because they don’t get any feedback from their auto-drummer. They need to be prepared to be criticized for their poor use of a musical shortcut.
I hope groups that use canned drums do record with live drums in the future, as they find the time or funds, or simply come to the realization that the active collaboration far outweighs the perceived benefits of a canned drummer.
In the same vein(s), I’m going to be disappointed if, for example, the video producer that Igorrr uses continues to only use an AI palette moving forward, and I think it would lead to creative bankruptcy pretty quickly. But if it’s a single item in a larger toolbelt? It’s a pill I can swallow every so often if used in the hands of folks who know what they’re doing. We’re never going to wish this medium away, but I hope we can strictly relegate it to the category of “fortifying pre-existing art” vs. “Acting like it can replace artists whole-cloth.”
Reminds me of Aphex Twin and Squarepusher a bit. I know glitchy, experimental music like this is not everyone's cup of tea, but I love it. It makes me feel kinda weird listening to it, like I'm on something. :P The bizarre AI generated music video actually suits it really well too, IMO.
Igorrr! Love him. I'm indifferent to the video portion of this, but the music is great, makes me think the new album will be quite promising. 'Savage Sinusoid' is one of my all-time favorite albums, but I felt 'Spirituality and Distortion' was a weaker effort. I hope this new one leans into the Squarepusher-y breakbeat type of stuff that dominates the first half of this song.
I’ve enjoyed all his releases, but I’ll always have a soft spot in my heart for “Hallelujah,” primarily because it was my introduction to Igorrr. How does that one stand up to the others in your mind?
Better than S&D, but not quite as good as 'Sinusoid'. I suspect his first couple of releases were a process of trying to figure out how to fit everything together - which is hard when 'everything' consists of black metal, breakcore, classical music, French folk music, up to and including the proverbial kitchen sink. 'Hallelujah' has a few really great tracks but a lot of it still sounds slightly arbitrary, like it's asking 'what could I put here' as opposed to 'what should I put here'.
Sinusoid, on the other hand, feels more in touch with the structural or even 'music theoretical' aspects of each of its parts, so that even though on the surface they may seem arbitrary (chiptune synths over blast beats?) there is a deeper connection that creates a surprisingly consistent aesthetic. Actually, I feel like S&D does about as well as Sinusoid in this respect, but at the same time it was more restrained or even formulaic with its approach. And fundamentally, I'd rather an album take big swings (even if they're not all winners) than play it safe - though of course with Igorrr, 'safe' is a relative word.
Haha that's fair; I think the kitchen-sink approach is what I liked so much about "Hallelujah," though its cohesion definitely does take a hit because of it.
I've dug your recommendations in the past, so I'll pose another question if you'll humor me - do you know of any other acts that scratch the same itch that Igorrr does? It's a very specific itch, granted :) you did mention Squarepusher and I've really only checked out one of their albums; they're probably worth a revisit at the very least.
rubs hands gleefully
Okay, let's see... It's hard to find anything exactly the same as Igorrr due to how unique his approach is. But what you've labeled an 'Igorrr itch' might more generally be for music that 1) blends lots of styles, focusing especially on electronic/breakbeat and metal, 2) is really fast and complicated, with so many ideas rushing at you so quickly that it's hard to keep up, though if you manage to, the feeling is exhilarating, and 3) has a sense of humor, plays 'musical jokes' on you.
Items 2) and 3) are actually more related than you might think. The essence of comedy is the unexpected, and the more syncopated and faster music is, the harder it is to expect the next note. Yet well-written music may also have the overarching logical structure necessary for humor (what in jokes often boils down to double meanings). I often find myself laughing at fast, clever music - I remember hearing Archspire's 'Relentless Mutation' for the first time with some friends, and we were all cracking up (that might actually be a good recommendation, come to think of it, though it's more purely metal than Igorrr). I also remember a few people laughing at a show for Squarepusher's live band, Shobaleader One, for more or less the same reasons, I imagine.
And indeed, Squarepusher might scratch this itch pretty well, but he has a big, diverse discography, so selection is important. His early releases may have the requisite speed and complexity, but by today's standards they're kind of conventional, and the timbral palette is probably not diverse enough. I think the first really 'way out there' Squarepusher album, and my personal favorite, is 'Ultravisitor'. Certain tracks might scratch the itch, but there might be too many slow, almost ambient ones in between. So maybe a better suggestion is 'Ufabulum', which is more uniformly high-intensity, but maintains a really sophisticated and experimental timbral approach. It's purely electronic, though I feel like certain tracks have a metal-ish vibe to them. 'Damogen Furies' might also be a good pick - quite similar to 'Ufabulum', perhaps even more intense, though also less nuanced.
Okay, on to other recommendations - check out 'Power Tools' by Big Lad. It's a fusion of grindy powerviolence style hardcore and breakcore or gabber electronic elements. Super fun, and has a way of, just when it feels like its intensity is peaking, getting more intense. No vocals though. A similar pick to this might be drumcorps, who has a bunch of great EPs, my favorite being 'Better Days' - a mathcore/metalcore/breakcore fusion with digitally fucked-with harsh vocals - lots of over-the-top fun. And then there is Machine Girl, who uses a lot more synths than these other two, and whose songwriting is a little more straightforward, but still brings a lot of metal tropes and is a great, somewhat irreverent vocalist imo. 'U-Void Synthesizer' is my favorite of theirs.
There are also quite a few electronic artists who don't have much of a metal influence, but still have intense, fast, and inventive songwriting, along with stylistic diversity, which I think could fit the bill. Iglooghost's latest releases have been a bit milder, but their breakout album 'Neo Wax Bloom' is a great piece of future bass/IDM with tons of energy and overflowing creativity. Lynyn is the solo project of the guitarist from the band Monobody, and his album 'lexicon' is really good, very textural in a way that's reminiscent of Squarepusher but also more modern. A left-field pick might be Gazelle Twin, whose album 'Pastoral' is really incredible - a mix of British folk music, creepy dark electronica, and aggressive noise. It's definitely way less intense than Igorrr on the whole, but might make up for it with sheer weirdness and, at times, dark humor.
Oh, and I almost forgot MIRAR's 'Ascension' - this just came out and I'm still digesting it, but it's this very unusual mixture of sludge metal riffs, dubstep's screechy noises and paced out rhythms, a little drum 'n' bass, flavor and then an out-of-nowhere dose of classical music. I think maybe on paper this might be the closest to Igorrr, but I'm not quite sure it meets the mark quality-wise, at least compared to my personal opinions on the rest of these releases.
Really, nothing's going to compare exactly to Igorrr, but hopefully you find something you enjoy here. Let me know what you think! If you like something in particular I might be able to provide more targeted recommendations.
Love this; thanks as always for sharing!
You brought up a great point about the musical surprises making one laugh - I actually think I’ve had the exact same experience with Archspire myself, I totally get it!
Outside of that act, I’m also familiar with Mirar’s work too. On the whole, I preferred their Mare EP to Ascension, though I still enjoyed Ascension and do need to spend more time with it. Mirar’s work is all “thall-adjacent” and is heavily influenced by groups like Vildhjarta and Humanity’s Last Breath, two other groups that I really like that definitely have a lot of the same aggressive approach that can veer into a different direction unexpectedly and make you laugh. Big fan!
On the more electronic side, I enjoy Venetian Snares for much the same reason - breakcore is already chaotic, and he takes it to that next level while still providing musical “through-lines” that keep his pieces from ever descending into pure noise.
Outside of those couple of groups, virtually everything else you’ve mentioned is new to me, so I’ve got some work to do. I’ll poke around and be sure to keep you posted; thanks again!
I was going to respond to the @eyechoirs comment about including ol' Venetian of the Snares, but didn't think it quite fit in their fantastic suggestions. Happy you brought them up! Venetian Snares is up there as one of my favorite artists, period.
VSnares is great! I probably should have mentioned Rossz Csillag Alatt Született (which might be my favorite album of his) due to the heavy use of classical music samples, which I guess is adjacent to Igorrr.
I saw this link, then watched the video and saw "Meat Dept." I was immediately stoked, I saw their VERY NOISE video which led me to their short film Black Holes, which is a pretty great animation.
This has their style all over it, assisted by whatever they did with genAI. From their description there seems to be initial 3D renders enhanced by genAI, which is also sort of shown by the design they use (there's sort of an exaggeration you can see in Black Holes that's present here). I thought it was an interesting video, and thought the sort of soulless AI vibe helped with the content a bit.
Didn't really connect with this and found the AI use off putting, but maybe not the way they intended.
They're doing... Something.
Well that was grating. I assume it's supposed to be with the visuals?
Yeah, I mean this is just kind of how Igorrr is. It's definitely not for everyone, lol.
Not specifically, Igorrr makes Metal-Focused IDM/Breakcore ("Intelligent Dance Music", a tongue in cheek response to "EDM"). The music is certainly made to be on its own, they just also make some really fantastic videos (this one uses AI + Traditional Rendering, whereas the others use just traditional rendering).
It's certainly a type of genre that isn't for everyone, personally it's one of my favorites, but its chaotic and very different than anything else here. It's the breakbeat, drum focused, glitched version of freeform jazz, essentially.
Still has that uncanny-valley effect and visible artifacting, but I suppose it works better for their style; the last video of theirs that I've seen, VERY NOISE, was very, uh, interesting.
With the title of "ADHD", as someone with severe ADHD, I was curious.
I'm still confused as to what exactly the message is supposed to be. Although with projects like this, I know often they will leave it up in the air a bit. I could definitely see some possible "ADHD is overdiagnosed and medications turn people crazy", but also perhaps just some representations of what ADHD can feel like to have. i.e., a negative and positive approach to the messaging.
I don't know. I have mixed feelings about what's being portrayed, but my instinct is that it is not positive, which is annoying. I deal with enough negativity in my life thanks to ADHD and to people not understanding ADHD that it's always annoying to see this sort of thing. Like how, as a geek, seeing the decades of movies and shows with the trope of nerdy geeks being socially awkward and weird doesn't reflect the reality that - yes, we can sometimes or even often be socially awkward and/or weird, but most people can't identify geeks by sight like you can in a movie. That trope gets… tiring. In the same way, ADHD tropes are also tiring.
I don't require the conversation about ADHD to be positive. It is, in fact, a neurological disability, and while I'm not totally unhappy having it; while there are positives and things about it that help make me who I am; at the same time, I have paid so much ADHD tax in my life, and I will probably die sooner than I would otherwise thanks to ADHD. That being said, I do appreciate the positive conversations about ADHD - how to understand those of us that have it; how we can understand the neurotypical folks; learning coping techniques to minimize the negative impacts.
Meanwhile, on the music from this artist… hadn't heard of them before, not sure I'll go out of my way to find them. I hardly considered this "music" — that said, that's personal preference I'm expressing. I can recognize and appreciate the technical expertise here. They are good at composition, and if this style is up your alley, this seems to be a good example of it. So while it's not for me, it seemed pretty good otherwise.
The AI was an interesting use. I think what interested me most was the ability to have recurring characters. And the technical skill to be able to get something that they wanted rendered. I can imagine this took a LOT of work. Pulling numbers completely from my ass, I would imagine that they surely had to generate a lot of clips to get ones that had what they wanted. It's rather impressive. And from a visual standpoint - the music of the eyes - it was visually appealing. I really liked the colour grading, for example. It really had that 50s era feel. I'm not against seeing more AI. It's useful already for surreal projects like this. I can't imagine what another 10-15-20 years will bring us.
It reminds me of how people have been against recorded media - records, movies - because you can't beat live performances. But movies and albums have been wildly successful. I think we will enter an era where we still have movies; still have live theatre; but also have AI-generated media that will allow for storytelling that is truly amazing. To allow my ADHD mind to follow yet another rabbit hole: People claim to hate CGI. It is my humble opinion that what most people hate is bad CGI. Good CGI? You don't notice it. Like good practical effects. So consider once we get to the point of AI being able to generate stuff that is longer, with dialogue, to the point where you can't tell it apart from live actors… we'll be able to have shows without the "alien of the week" problem, i.e. aliens who have to be in relatively human-sized costumes because humans play them. We can have truly alien beings interacting with humans because it's all generated.
I think there will be room for that, along with more traditional movies with CGI and things like live theatre, if that makes sense.
Gautier Sierre (Igorrr) has a note on the video about the song
Essentially, the goal is to express disjoint, loosely connected passages similar to how one might have issues focusing on a single thing with ADHD. From his description it's autobiographical, which may help contextualize it a bit for you.