Hey everyone,
Lately, I've been seeing a lot of editors experimenting with AI tools — from automatic beat-syncing to AI-generated motion masks and even entire background replacements. It made me wonder: where do we draw the line between creative assistance and losing the human touch that makes AMVs so special?
Would you consider an AMV made partly with AI effects (like Stable Diffusion scenes, AI rotoscoping, or voice isolation) eligible for contests or VCAs?
Or do you think that kind of editing breaks the spirit of what AMVs have always been — raw creativity, rhythm, and editing skill?
I’m really curious how everyone feels about this — especially veteran editors who've seen the community evolve over the years.
Have any of you tried using AI tools yet? If so, what’s been your experience?
Let's talk about the future of AMVs — is AI the next big tool… or the beginning of the end?
Should AI Be Allowed in AMV Contests?
- Rider4Z
- The Machine
- Joined: Sun Dec 14, 2003 3:55 am
- Status: Larger than life.
- Contact:
Re: Should AI Be Allowed in AMV Contests?
Many contests are beginning to implement AI rules, while others don't care either way.
For Sakuracon, our TL;DR version is no generative AI, but you can use it as a tool for cleaning your source materials.
For Sakuracon, our TL;DR version is no generative AI, but you can use it as a tool for cleaning your source materials.
- Zarxrax
- Joined: Sun Apr 01, 2001 6:37 pm
- Contact:
Re: Should AI Be Allowed in AMV Contests?
I've got some thoughts on the matter, but like any big issue, it's complicated. While I would generally consider myself to be in favor of AI, I recognize that it brings problems along with it, and I can understand reasons why AMV contests may be opposed to some forms of it or find it problematic.
First, the positives. AMVs to me have always been a kind of creative outlet. I imagine something in my head, and then I want to bring that imagination to life. The actual result is usually compromised in some form or other, but it's exciting enough that I am still here after all these years. I see some of these AI tools as an incredible step forward in helping me bring my creative vision to life. They allow for things that I could have never done just a few years ago. The biggest problem that I run into with making AMVs has always been the restraints of the original source footage. I can't even count how many times I have simply had to give up on a video because footage that I wanted to exist simply didn't exist. In fact, just this week I had to give up on an idea that I had been dreaming about for months. But, I could make it happen with AI if I wanted to. So basically for me, I see these tools as removing restraints and letting me achieve my creative vision more freely. I can recall over twenty years ago being so excited when I saw a research demo of an automatic rotoscoping technique, and dreaming about the day when I would be able to use that in an AMV. That day is finally here.
AMV Contests can have all kinds of rules about what you can and can't do in your videos. Some contests might not want any audio from English dubs, or some might be heavy handed in censoring things that might not be family friendly. I think contest coordinators are trying their best to put on a good contest while weighing the needs of different stakeholders and making sure that things are fun and competitive. AI tools are probably going to make their jobs a lot harder in the short term though.
I predict that within 5 years, trying to restrict the use of AI is going to become a mess. We are going to see AI being used by animation studios, making it really hard to restrict when it's already burned into your footage. Editing software will continue adding more AI tools, blurring the lines between which tools may or may not be acceptable to use. Younger editors are going to be growing up around this technology, and may end up not really being able to differentiate what's AI assisted and what isn't. Do you know how the engine in your car works? Should you be expected to know the difference between how "content aware fill" and "generative fill" work?
And just think of images. AMV editors have forever just grabbed images off of the Internet and slapped them into their videos. Now, there is a very strong chance that when you find an image on Google, it has been generated or enhanced by AI. And you may not even be able to tell. How is that going to be restricted if the editor didn't even realize it?
When it comes to specific tools, I have not yet seen any pushback on AMV editors using masking/rotoscoping or voice isolation. Video upscaling and cleanup also seems to be fairly well accepted.
I've been using AI in my videos for several years now in one form or another. I've found voice isolation to be particularly useful at times. I have also used upscaling a fair bit (I have even trained my own upscaling and restoration models). I've only used rotoscoping a small bit despite developing my own tool to do it. My most recent video uses image generation and video generation to achieve some effects that I could not have done any other way. I'm also currently running up into the issue where I may have to make both an AI and Non-AI version of videos. It feels kind of bad to have to send what I consider the "lesser" version of a video into a contest, but it is what it is.
I have been putting together a document to list some various uses of AI, which contest coordinators or editors might find helpful. (For example, to help identify which tools or techniques might be considered acceptable or not acceptable)
https://docs.google.com/document/d/1Pnr ... p=drivesdk
First, the positives. AMVs to me have always been a kind of creative outlet. I imagine something in my head, and then I want to bring that imagination to life. The actual result is usually compromised in some form or other, but it's exciting enough that I am still here after all these years. I see some of these AI tools as an incredible step forward in helping me bring my creative vision to life. They allow for things that I could have never done just a few years ago. The biggest problem that I run into with making AMVs has always been the restraints of the original source footage. I can't even count how many times I have simply had to give up on a video because footage that I wanted to exist simply didn't exist. In fact, just this week I had to give up on an idea that I had been dreaming about for months. But, I could make it happen with AI if I wanted to. So basically for me, I see these tools as removing restraints and letting me achieve my creative vision more freely. I can recall over twenty years ago being so excited when I saw a research demo of an automatic rotoscoping technique, and dreaming about the day when I would be able to use that in an AMV. That day is finally here.
AMV Contests can have all kinds of rules about what you can and can't do in your videos. Some contests might not want any audio from English dubs, or some might be heavy handed in censoring things that might not be family friendly. I think contest coordinators are trying their best to put on a good contest while weighing the needs of different stakeholders and making sure that things are fun and competitive. AI tools are probably going to make their jobs a lot harder in the short term though.
I predict that within 5 years, trying to restrict the use of AI is going to become a mess. We are going to see AI being used by animation studios, making it really hard to restrict when it's already burned into your footage. Editing software will continue adding more AI tools, blurring the lines between which tools may or may not be acceptable to use. Younger editors are going to be growing up around this technology, and may end up not really being able to differentiate what's AI assisted and what isn't. Do you know how the engine in your car works? Should you be expected to know the difference between how "content aware fill" and "generative fill" work?
And just think of images. AMV editors have forever just grabbed images off of the Internet and slapped them into their videos. Now, there is a very strong chance that when you find an image on Google, it has been generated or enhanced by AI. And you may not even be able to tell. How is that going to be restricted if the editor didn't even realize it?
When it comes to specific tools, I have not yet seen any pushback on AMV editors using masking/rotoscoping or voice isolation. Video upscaling and cleanup also seems to be fairly well accepted.
I've been using AI in my videos for several years now in one form or another. I've found voice isolation to be particularly useful at times. I have also used upscaling a fair bit (I have even trained my own upscaling and restoration models). I've only used rotoscoping a small bit despite developing my own tool to do it. My most recent video uses image generation and video generation to achieve some effects that I could not have done any other way. I'm also currently running up into the issue where I may have to make both an AI and Non-AI version of videos. It feels kind of bad to have to send what I consider the "lesser" version of a video into a contest, but it is what it is.
I have been putting together a document to list some various uses of AI, which contest coordinators or editors might find helpful. (For example, to help identify which tools or techniques might be considered acceptable or not acceptable)
https://docs.google.com/document/d/1Pnr ... p=drivesdk
-
- Joined: Sun Oct 12, 2025 10:05 pm
Re: Should AI Be Allowed in AMV Contests?
Great topic — and a really important one right nowgordonpiper wrote: ↑Sun Oct 05, 2025 9:38 amHey everyone, Slope Run
Lately, I've been seeing a lot of editors experimenting with AI tools — from automatic beat-syncing to AI-generated motion masks and even entire background replacements. It made me wonder: where do we draw the line between creative assistance and losing the human touch that makes AMVs so special?
Would you consider an AMV made partly with AI effects (like Stable Diffusion scenes, AI rotoscoping, or voice isolation) eligible for contests or VCAs?
Or do you think that kind of editing breaks the spirit of what AMVs have always been — raw creativity, rhythm, and editing skill?
I’m really curious how everyone feels about this — especially veteran editors who've seen the community evolve over the years.
Have any of you tried using AI tools yet? If so, what’s been your experience?
Let's talk about the future of AMVs — is AI the next big tool… or the beginning of the end?