Implications of AI for Video Editors | Cutting Edge: Editors on AI | Adobe Video
Hey, everybody. My name is Megan Keene, and I'm part of the pro video team at Adobe. But I actually started my career as a documentary filmmaker. I spent almost a decade making documentary films, and I've been at Adobe for almost 13 years. And what keeps me continually excited about working in this industry is that video is always evolving. Most recently on the scene is AI.
It's the big topic of conversation. And along with that topic comes a lot of considerations. We need to be thinking about how we're forging this new path forward. And that's why we're here today, to be having conversations with filmmakers, with creators, to really talk about what should we all be thinking about as we're forging this new path with AI? And with that, I'm very excited to introduce my friend Stacey Gold, who is here with us today. Stacy, thank you so much for being here.
Thank you for having me as by way of introduction, maybe you could share a little bit of some projects that people might be familiar with, and a little bit about how you got started in filmmaking. I have edited a wide variety of documentaries, many in the music space, documentaries about difficult subject matters, difficult conversations, political documentaries, true crime documentaries. I mean, it's run the gamut.
And so how did you get your start in filmmaking? It actually goes back to the days of being a college radio DJ in Nashville, Tennessee, in the 90s. and so I actually had a radio show where I would go out with my little Marantz audio recorder and go out and interview people in different locations. And I discovered that one of my favorite things is being I love being a fly on the wall. I love having the opportunity to go into spaces I would never normally be invited into, and just be able to listen and learn and be curious and hear people's stories. And so that's kind of how it started, was editing little radio documentaries. And I feel like your description of that is literally the definition of being a documentary filmmaker, just like entering these worlds that you're not necessarily born into, but you get to be a part of it because of your, your storytelling.
You know, other times working on projects where you are a part of the community and you're learning a lot about yourself. Absolutely. So in this topic of I sort of what is your familiarity with AI now? Are you incorporating it into any of your workflows? How are you feeling about this new influx of AI in the filmmaking industry? One of the things that, about AI is it's so broad, it's everything.
And thinking about AI. I'm realizing I've been using it for so long that it's such a huge term, artificial intelligence. I mean, would we consider that I was using AI back when the first Macs came out, the Mac Classic, and we were playing, you know, video games. I mean, would that be considered artificial intelligence video games? I don't know where it begins.
It's funny because I think a lot of filmmakers when when we talk to them and ask this question, you know, are you using AI if they haven't really thought about it often, it's like, no, no, I'm not. And then there's this, lack of understanding of what features actually are. I mean, in premiere Pro alone, there have been features for almost a decade in Premiere Pro that are AI, that are machine learning or artificial intelligence powered. At Adobe, we think about AI in two different buckets. So there's like the assistive AI, which are often the features that people don't even realize they're using speech to text. Auto reframe, auto duck, all these things that really take the tedium out of your workflow.
More recently, is the generative AI topic right? It's the. Net new pixels. Net new wave form, sort of creating something out of an idea in your head. What's interesting is you and I are somewhat contemporaries and coming up in the industry around the same time we were there in that shift from film to digital. And I feel like a lot of the same topics that are coming up now were fears, concerns, considerations back then.
Can you think back to that time and tell me, like, what were people worried about when we were shifting from film to digital? I remember when I was in film school, I went to the school, the Art Institute of Chicago, and I was in the film program MFA program, and this really great experimental filmmaker name, my Isabelle Carpenter taught me, hand processing. So and that was something that she was doing was a small community of experimental filmmakers who were doing ham processing. One of her specialties was spaghetti style ham processing. So we would literally film on 16 millimeter black and white.
More was easier. And then take it into the photo department and process it like processing film, but do it spaghetti style where you rather than create put it on reels, meticulously use reels. We just shoved it all in like spaghetti war gloves, of course, and just went, you know, and just moved around physically touching the film. Then we would take the film out, you would hang it to dry, and you had this beautiful, beautiful film that you could see the work that you had done on the film. It had been touched, it had been hand processed. What was interesting was when I showed that footage at a critique, there were people in the room who were looking at my footage and they said, oh my God, what filter did you use? And I just was like, okay, no filter, but all of this, they just thought, oh my goodness.
What filter did you use to make this scratch scratch happen? Right? The big difference I feel in these transitions is that we actually had to learn it. Yeah, we had to consciously say, okay, I better learn final cut. I better learn I have I haven't taken a like when I first worked on premiere, I had somebody teach me premiere. I'm not inviting anybody to come over and like, oh, I need a tutorial.
And I so that I can work this job. At least not now or not yet. It's just integrated seamlessly, I feel, into our work that we're using it. We don't even realize it.
You know what I mean? Like I start, oh, I can do a search now for in the script for this, which we've been doing for forever, but. Well, for not forever, but for a while. I didn't need somebody to teach me that.
I just found it, you know? So to me, that's what's different. There's also I mean, I remember there being a lot of conversation around, oh, video's never going to look as good as film. Yeah, it's never the quality is never going to be as good or there was a lot of concern around job loss.
Right? Right. If the computers start doing everything right, what happens to the people, which I think is a lot of what we're considering in this. I mean. There was job loss. I mean, negative Cutter's there's a lot of people in the film in the film industry who are gone from conformist film labs. So there was job loss. And not to mention, like we said, the people who just they felt like they couldn't adapt.
I do think there there can be job loss. At the same time, there's new jobs, right? Job. So I think it's one of those things where you have to adapt. That's one of the biggest topics that needs to be talked about is so often. And I remember when I first approached you about this conversation and you were like, if I'm honest, like, I haven't been paying enough attention.
And I think a lot of people's inclination is to be like, well, I just I'm just going to keep doing what I'm doing. And like, that might be happening, but I think it's really important to actually participate in the discourse, to actually keep yourself informed about what's going on so that you're making educated decisions, especially for someone like you, where you have the power to say like, no, I need my assistant, right? I'm not going to just say like, oh, I'm just going to let I do all my organization and I can let the producer start scratching out that line item of the creative. So I think there's a couple things I would love to talk about. There's process, there's efficiency. There's the question of, I can do for you what you least like to do. So I think there's like a couple different conversations we can have here.
I think if we think about AI, generally speaking, when when you're as an editor, a lot of times we're constantly balancing. When we're professional editors, we're constantly in this battle of creativity versus efficiency and making your deadlines. I mean, we we live and die by our deadlines. How do I make art and be as creative as possible and feel that I am giving this project the time and the thoughtfulness it deserves, and make my deadlines and every editors different. We all have different superpowers.
We all have different tools. We. That's what's the beautiful thing. And one of the things I love about premiere, that I love about non-linear editing, is that there's many different ways to do something. No two people necessarily have to work the same way and the customizability of it. But for me, so much a part of what makes my work work is knowing my footage. It's all about process.
And so I have a lot of concerns about like content awareness, for instance. Now, not everybody likes to watch all of the footage, and not everybody feels they have time to watch all of the footage. I would argue that we are faster and more efficient and better editors when we know our footage one. Hundred percent. And I get a little concerned sometimes if an editor says they're not watching their footage or they're rushing through it, or they're just reading waveforms, even if we just read a transcript or, you know, because we lose a sense of how something was said totally about.
Emotion behind the. Emotion. But how many times do you look at something on a transcript and it just like is nothing, and it's everything, you know? Yep. and then also just expressions, moments, pauses, breaths. You know, I once worked on a documentary project where I noticed that, one of the main, characters in the documentary, her neck would get bright, bright pink red when she was uncomfortable.
And that was something that I notice. I don't think content awareness would necessarily catch that. Right. So if I say, I want you to find all of the reaction shots where this person looks uncomfortable, there's certain things they might be looking for. Are they looking for a redneck? I don't know, right.
And I wouldn't know to ask for that. So you and I have both worked on projects with so much footage that it's almost untenable. Like I'm thinking for you about Lollapalooza, right? Years and years and years of footage that you were working on. I worked on we live in public. We had thousands and thousands of hours of footage. I watched it all.
You watched it all, but then you go to be like, I remember that there was this person talking about this thing, and they were wearing this, and I remember them doing it, but like, where is that? How do I even go about trying to find that moment that I know exists? Because I watched it? But the the process of sort of finding it more efficiently, it's not necessarily don't watch it, but could it help you get to that moment that you know, you remember it being something like, oh, I definitely, but then where is it in the in the math? Okay. So not for me because I have a process, but for my assistant. Yes. So maybe because when we watch our footage we're not just eating popcorn watching our footage. My process, which is a lot of people's processes, your markers. And.
And you're writing it down, you're creating sequences, you're creating byte maps, you're creating selects. I would argue, and again, this is part of the conversation of efficiency and process for me. We know that we learn in many different ways. Some people learn best by seeing something. Some people learn better by hearing something. I learn best.
And a lot of people do by multiple ways. I hear it, I see it, I write it down, I process it, I go on my dry erase board and I write a thought down. I might write a card, I might put a card up by having a variety of ways to interact with the footage and have a relationship with the footage. That's not just one passive way of looking for something, watching something. I am developing a relationship with my footage and the best relationships that you can have with another human being with anything is multifaceted.
It's tactile. It involves thinking. It involves pausing. When I worked on disclosure, there were moments where somebody would say something so powerful I had to stop and I had to go for a walk because I just had to take it in. I needed the slowness. I needed the time to just take it in.
And then I marked it up. As a result, I knew where it was. I knew where that I just remembered. I remember who set it so I knew how to go into their bin. I knew how to find it.
The markers were there, the selects were there. I felt like right away. So it wasn't a problem. Where it might be helpful is if I have an assistant editor who doesn't have time to watch everything. Although I really like when they do too, because I like when there's a collaboration.
Or if they don't grasp the context that you may grasp when you watch. They see it a different way, they have a different relationship. And I might just be in a real hurry. And I say, can you just find me a shot? Any shot of somebody on a horse? Yeah, just can you. I just need a shot of a horse right now.
It would be better if I had an assistant editor do that for me. maybe an assistant editor can go through the footage as well. Maybe it's been marked up because I like when an assistant editors work is not just technical. It's a collaborative experience. Sometimes. It's often a story producer will do that.
I also think in especially in documentary editing, there are some things that you'll watch, and they seem insignificant in the moment that you're viewing the footage for the first time, and later on you're like, oh, wait. I remember there was a moment where the cat walks through the hall and then you know what I mean? Like there's like this moment that you don't even recognize until you start getting into the storytelling process, and then you're like, oh, there is something that I could use there that would, you know, be relevant that I hadn't thought of in the moment. Viewing it the first time. My mantra is when in doubt, market or when you're pulling selects. When in doubt, put it in where you get selective is. Then when you start putting together your kind or putting a rough cut, you know, and starting to build your strength out, then you have to be more selective.
And then if there is a doubt, then that's when you leave it on the cutting room floor. If you're in doubt, you're not sure it doesn't belong in the cut. I have really meticulous marking is an interesting moment happened that I remembered. I probably would have marked it.
And then if, you know, if we can search markers, if we can organize markers better, I use that function. Now. This is where I be very helpful. I have had situations where I've worked on a project, and the cinematographer watched a cut, or somebody who was in the field, or even the director, the producer, and they said, I remember this. I might not have seen it, then it might be helpful to find it.
So more to like. What are some of the trends or places in your current workflow where you're like, oh man, this is where I would really love I to to step in and take on some of this. A question that I hear happen a lot when people talk about using AI is what do you least like to do? What? What is something that you don't enjoy? So then let I do it for you. And I would argue in a lot of times with that, well, if you if I don't like to do it, let's pay somebody else to do it right. If I don't like it, that's why we have other jobs. Or I mean, I think sometimes those friction points or the pain points in your workflow force you to look at things a different way.
There's a cut that I like not being able to figure out, so I try a couple of different ways. It uncovers new possibilities or new creative paths. We can think of I as a collaborator, but not as a replacement. You know, I one of the things that I think happens when you're editing a film, whether it's documentary or it's scripted or it's television or as editors, we often have to wear a lot of hats that we know aren't necessarily trained for. When you're an editor, you have to understand music for sure. You really. Music is a very important part.
Sound design is really important. Understanding basic titles. You need your graphics team. You need your sound team.
You need your mixers. You need your sound designers. You need your composers. You need the whole team. You can't be all of that. And I cannot replace all of that. What it can help is how we as editors and also art the directors and producers, how we can communicate to the highly skilled and experienced people who are our partners on the project.
Like when it comes to fonts, for instance. To me, that's a waste of time to go through every single font that's listed and try each one. That's probably one thing that I could really do without, is the time I waste having to even put together like a slate putting together titles. Ideally, there is a graphics team that is coming in and that's their job. They're going to pick out the font and it's a creative conversation. What does that font represent? It's not just something that looks pretty, it's something that actually represents what the project is about.
I can recognize a beautiful font. I'm like, oh, great. When I start to try to do it myself, I'm like. Oh, there's people who love fonts.
It's an art form in of itself. I mean, there was a whole documentary made about Helvetica, you know, it's somebody else's passion. It's something that's somebody else's job.
But in the meantime, I can't be calling them up and having them come in while I'm in build mode. I'm not even at rough cut one. They're not. It's not time for them to come in yet.
Ideally, it'd be great if they could come on from the beginning and they could give us some samples and some ideas, but sometimes you just got to slap something together so you can get your cut going, and then the person appropriately steps in and says, okay, now that I've senior cut put together, I'm feeling this font. What's so interesting is that we know as editors, though, even though you have acknowledged fonts aren't my thing, it's not like where I want to spend my time. If you choose one and it stands out like a sore thumb, like your director, your producer is going to be like, wait a minute, that font isn't working.
You're like, oh no, it's not the final font, but like, let's move on. And that is, I mean, like in the conversation of efficiency, those things where, yes, it will go to sound, it will go to color, it will go to, you know, your graphics team being able to not have that break in the creative flow with your creative collaborators is so important. If you think of it this way, if I was really amazing with fonts like let's say I had my own font book and I created fonts on the side and I put together some incredible font in Rough Cut one, we would still have a graphic artist come in and say, okay, let's work with that. But now I'm going to add to that, I'm going to do this, and they're going to do what they get paid to do, working with a highly educated, collaborative team who has given them something to be inspired off of. So if I can help with that, might make their job easier, because they're actually dealing with people that already have a vision or a sense of what they're working with.
It gives a spark of like, this is the direction we're going now. You elevate it or evolve it. Yeah. I mean, as collaborators, we don't necessarily want to be given a blank slate.
Like when I work with directors and producers, I like working with people who already have a vision. They already have an idea. They put it all on me as the editor. Okay, like that's fine, but I it's more rewarding if you have a vision together.
We're coming up with something we never thought we could. So if I can help people who. That's not necessarily the language they speak.
You know, fonts come to the table to the graphics team. I think it could make for a more rewarding collaboration. I would say the same can go with sound. I worked on a music project once where we had the full library of of an artists, music, and there did exist stems for all of their music, so we were going to get the stems eventually, once the film was locked. You could have the vocals separate from the guitar, the drums or. Right, okay. Yeah.
So that they could mix it properly, but also so that in a documentary we can use their music as the score for the film. Well, eventually we were going to get all of those stems, but we couldn't request stems for every single song in their vast collection. Nobody had the time or the money or the resources to give us stems. It would take in somebody a lot of work, and it would have been too much. So to be able to temp it out and be able to use a AI to create temporary stems is something that could be very helpful for people working at the music documentary space, and it doesn't replace the work of the music editor. It doesn't replace the work of the mixer because it's just a temporary fix.
There is an art to how you use sense to how you mix the stems. That is somebody else's job, but it's just a way that you can figure out. What songs are we using in this film? Okay, these are the specific songs we can ask the rights holders to send us stems for.
So that's an example, I think, where AI is a collaboration and it helps us create proof of concepts so that you can get your cut to a place where people can see it. Most people don't know how to watch a rough cut. Even editors. Right? Good point. Yeah. Whenever somebody says to me, oh, just send it rough, I can watch a rough cut. I'm like, no, no, no, no. It's rare. Sometimes a director producer can and that's a gift.
But a lot of times they can't. And as editors, we're only as good as our last cut. If we turn something in that. Oh, no, just send me something rough. I really.
And it's not beautifully finessed, and it doesn't have the music and everything's just right. You might not have a job the next day, so it's important for us to be able to show our vision and to show what we see as finessed and beautiful as we can for that first screening. So as we think about how we are building AI, as we are thinking about the needs of filmmakers in the filmmaking industry, how do we ensure that AI is built responsibly? What are some of the considerations that you think should be top of mind in this conversation? There's like a mantra that I think has been coming out of, like SAG has been using in some of their negotiations around AI. They talk about transparency, consent and compensation. I think about transparency.
There have already been some examples of documentaries where maybe they used AI to voice something or whatnot. And I think the question wasn't necessarily was that ethical? It was why wasn't it revealed? There's a film that I edited, directed by Lucy Jordan, called Our Father. We used heavy recreations, but all of the audio taped conversations, like voicemail messages and stuff, were real. At the beginning of the film, we said, this film uses recreations. All of the audio recordings are real.
We made it clear. So now people understand what they're dealing with. And we knew because of AI that people would think we were faking the recordings, which were such an important part of that documentary.
So I think transparency is really important. I think people, if they know what they're in for and they know what's happening, they're going to be less likely to be angry. Another thing I think we can talk about is like issues around consent. I'm thinking about how there have been some uses of generative AI, like in a film like Welcome to Chechnya, which protected people and journalistically protected people who would not be safe being seen on camera. That documentary by David France.
It's a really powerful documentary because we're seeing people, and then it was so powerful at the end of Spoiler Alert, but at the end of the film, they reveal it and you understand the power, the mere fact that their faces had to be changed. You know, that in and of itself is very powerful. Before this technology, their faces would have been blurred or they would have been in like a like, you know, un lit with. Yeah, yeah, yeah. So I think that's a powerful use.
It also can be dangerous because if you think that you're doing it but you're not, you might not culturally understand certain markers that can reveal who a person is and it can put someone in serious jeopardy. But I would say when it comes to jobs, a big conversation that we can be having right now is what are people's jobs? Because editors, we get disrespected a lot where somebody like, oh, I could do that, you know, or, you know, it's just technical. You know, how many times somebody just say, oh, you're an editor, what do you just do? Snip different? You know, it's like, no, we do a lot more than that. We're writers. We're storytellers. It's not just a technical job. It's so much more than that.
And I would say every division in the film industry could have the same conversation. For instance, someone who works in hair and makeup might think, oh, they can just use AI to change the way a person looks or creates special effects. But let's talk about that. Job is so much more than that. It's a part of the process of interacting with the cast before they go on set. Their job is so much more than just makeup.
It's part of the collaboration. You know, I think assistant editors, their work is so much more than grouping and syncing. Cinematographers don't just turn a camera on and it goes, you know, so much of the work that people do is thinking, and it's having a point of view that they bring to the set and seeing things from their unique perspective. So if we eliminate positions and we think I can do everything, we're losing that collaborative spirit of different perspectives.
Especially in editorial. The assistant editor position seems to be the most precarious in this conversation. People really there's a lot of concern around the assistant editor.
For good reason. I think partially there's been an evolution of the assistant editor role that isn't necessarily what any assistant editor, meaning somebody who wants to be an editor later on that there's like data wrangling, this sort of media management piece that, yes, some parts are very important in terms of organization and knowing the footage, but when it becomes like literally just like, you know, drive space in the cloud or down way like that, like this stuff that you don't necessarily know creative person really wants to be spending that much time on could be taken over by then. The question becomes, what is your responsibility as an editor to say, I'm going to take back that mentorship relationship with my assistant, that showing an assistant how storytelling can be done or giving them the opportunity to be telling more parts of the story or building assemblies or like, what ownership do you feel for that piece of the conversation? Well, first of all, I think with assistant editors, a big part of their job is to fix all the mistakes that everybody else makes. So assistant editor is never going to go away because somebody has to monitor what the AI is doing, has to double check, make sure things are really in sync.
A lot of times there's user error that happens that the assistant editors have to fix their, you know, there's all kinds of things that go wrong during production or in post in the transferring process. So I think a lot of assistant editors would probably appreciate tools that just make the tedious part of their work easier. But yeah, the originally assistant editors I was at the tail end of this, but the technical part was just minor.
But really, an assistant editor is an assistant to the editor. It is a creative position. That is also, if they want to become editors, is training them to become editors. I'm on a project right now where I always like to have conversations with the assistant editors I work with, and they say, okay, so what do you really want to do? What's your aspiration? Do you want to stay assisting, and do you want to go more into this or that? And if they say, I really like editing and or sometimes they want to produce whatever they want to do and then I'm like, okay, well then let's think of a way that you can support me in this way and do something that's enjoyable to you.
So I'm working on a project right now with an assistant editor who's having to wrangle a lot of footage and, you know, and just do a lot of technical work. But one day, on our group chat posed a very interesting question that completely transformed, like a big part of how I saw the episode, and she was absolutely the person to say what she said. She had a close relationship to the subject matter, and I was like, we have this person here who is so busy grouping and syncing, and we are missing a very important voice.
on this project. Do you foresee knowing AI tools or knowing how different AI technology can help in post-production will start to be a sort of feather in the cap of an assistant editor? Oh yeah. I mean, I think like they, you know, the question of what do you least like to do? I would say the question, what do you need the most support with? It's been so long since I went as an assistant editor that I'm not up to date on a lot of things, so I like having an assistant who balances me out. You know, I have an assistant editor. I've worked with for years, Tony Dang, who? I always joke about how I just want him to, like, spend a day with me in my life because he's always like, you know, there's a faster way to do this.
You know, the. Efficiency when you're doing this wrong. Oh, you know, everybody needs someone like Tony in their life because Tony is the one who can say to me, Stacy, you know, I know you're doing this, but you know, there's this tool that can, you know, and I'm all ears, you know? So I think as a collaborator, whether it's like my other assistant editor, Kelly, who came up with this great idea because she's also, you know, a thinker, a journalist, you know, and just has all of these other skills. And she can bring something. And so there's Tony, who just is like a whiz with all things tech.
So it's just having more voices in the room. One thing that I love about you is that you're very discerning. Like you clearly think about things a lot. Is efficiency the number is like speeding up how you work.
The number one thing that you feel like I could bring. Well, I'm sure that there's a lot of people who are like, bring it on if it makes you faster. So this is the thing, everybody's different.
But as you can probably tell, when I get excited about something, my hands are moving. I start talking really fast, and sometimes it's hard for people to keep up because I'm just like. So for me personally, I benefit from slowing down. So if I need to sit on something, I need to chew on things.
I get a great idea, I write it down and I like to think about it. I might want to process it if I'm moving too fast. I think the work could be better if there was enough time to process it and really build on an idea.
So speed doesn't help me because I'm already a very energetic person. I always argue, I edit faster because I know my footage. I get concerned if someone's too fast. Because here's the other thing. I think that if something comes to you very quickly, sometimes you have to interrogate that.
Is this an intuitive thing? And it's coming quickly because it's meant to be. And this is like, you know, the film gods saying, you know, this is the song to use or this is the bite to use. But sometimes when things come really fast, it's because it's the easiest route. I've been doing this for so long that sometimes if I make an edit, I'll take a moment and be like, well, that's the obvious edit. That's the edit I always make. That's the predictable edit. And sometimes I want to stop and say, maybe I should do something that makes me a little more uncomfortable, or something that's a little less expected.
It's easier, and it's faster to go with the go to and to work with the same toolkit. But sometimes slowing down and getting out of your comfort zone and trying something that takes that's harder. Failure is the friend of art.
We need room to fail. Now. If efficiency can help, take care of other things so that we and it creates more room for failure and it creates more room for experimentation, and I'm all for it.
Tell me about creative flow. What does that mean to you? Unfortunately, it's when you forget to stretch or take care of yourself. It's like years of back pain. I think you just lose yourself.
You just forget you're there. You're just in a groove. I mean, it's the same thing that happens with athletes, you know? It happens with anybody. When you're working.
You just forget that you're working. You you're just there. You're in it. You're connected with it. It's like dancing. You talked a little bit about the topic of transparency. I will venture to say it's almost more important in the documentary genre than anything.
Yeah, but we know as documentary editors that often you're taking sound bite from one interview and pairing it with a sound bite from a different interview. And like, they may not have been talking about that exact same thing both times, and then it turns into the the ethics and the morals of the editor and the filmmakers of like, are we really representing truth? Are we really representing the reality of this story? I'd love to hear sort of your thoughts in terms of the impact of AI and transparency on documentary in particular. I wish I could remember who said this, but somebody said, you shouldn't be learning your history from documentaries. You know, like there's a place for school and books and textbooks and lectures. You know, anybody who thinks that when you watch a documentary, you're an expert on something because you watch Ken Burns's documentary on the Civil War or something.
It's like, it's helpful. Documentaries definitely give us information. They definitely help us, but they don't.
They shouldn't replace other forms of education. They are a great tool, but we need to understand that any time something's edited, there's manipulation. Editing is manipulation.
When you add a pause that wasn't there, that's a form of manipulation. You know, when you lift words for seasickness. Is this in the spirit of what that person said? Well, even the role that you choose to cover. All that you do an. Edit is like you can make really impactful choices just with like what could be very benign B-roll, but which you choose can actually shift the audience's perspective of how they're, you know, digesting that information 100. Percent. Every choice, every frame has an opinion.
There's nothing like objective about filmmaking and about editing. I do think, that being said, that we have to really understand all the choices that we make, from casting to the questions that are asked to who asks the questions. You know, I could go on and on and on. There's so much manipulation and dishonesty that has been happening long before these conversations about I started. So then the question becomes, well, who are the people making the documentary and what are their intentions? And we all know what happens.
The road paved by good intentions, right? Even people with good intentions can do really stupid things and unethical things because for whatever reason. And so I think we have to have as many stopgaps as we can and have as many people on the team and have these conversations. And that's where being really fast gives you a little bit less time to sometimes question.
There are some things happening right now, like around generative AI, that are causing a lot of conversations. I know the Archival Producers Alliance for APA are having a lot of conversations about using generative B-roll for documentaries like, oh, I need a shot out of something and I can't be there, so let's just generate it like, that's terrifying because I well, for one, I mean, it's like it's not real. So I guess we need to say it, it was used. So I think that's where transparency would be important.
There's so much that's part of the process of searching for archival. It's not just about finding the perfect archival footage, it's about understanding when you're looking for something. So if an archival producer is looking for something, something else comes up in that footage that they didn't even know was there, right? Certain images are iconic. And also we don't know what we don't know.
I mean, to me, that's like a mantra. I, we we use that a lot when we're working on disclosure. We don't know what we don't know.
So you might not realize that there's something in that shot that absolutely says a different story. Then it goes back to the conversation we were having earlier about watching all of the original capture footage. Right. That's one thing.
But then the archive, like, who knows what's there, who knows what you might start uncovering as you start watching through the archives from different sources and you start saying, oh, well, you know, I'm watching this newsreel and I see that there's this other camera from this other news channel, like, let's go see what they got. And then, yeah, I mean, it's some of the most fun parts of documentary is like uncovering the rabbit. Hole, sweeping the sleuthing, the rabbit holes.
That's such an input. It's the process. And if we start cutting corners in the process, what are we doing? You know that it's so much of what the joy is of if you don't love looking at footage and you don't want to listen to interviews or, you know, read the transcripts or understand what you're working with, why are you in this field? Like, why are you doing this at all? If the process is so much a part of creation of art and you never know what's going to come out if we just plug everything in and we say, build me a film that has this, this, this, and has this kind of person, it's like, well, you knew what kind of questions to ask, but so much of it is the the accidents. Now, I know there's artists, I collaborative artists that say that is happening and I and it's my naivete that people are saying AI art is art, you know, and that there are accidents that happen and that's fine. But just say you're doing it.
Yeah. You know, I'm not saying that we should just cut off AI art. I'm not saying it's not art. I'm saying own it. You know, talk about it. So what advice would you give to young editors who are wanting to sort of get into the field? One of the most important qualities of any artist, and of another, is to be curious and to live your life outside of the space of editing. My advice to emerging editors is to understand that editing is so much more than editing.
Everything that we live, everything that we experience, everything that we see, every person we talk to, we bring into the editing room with us. And why are these conversations important to be? Having the discourse around editing and the craft and new technology innovation, and how it's being built? We are storytellers and storytellers fuel society and culture, and how society sees itself and the stories that are told, stories change people's hearts, change people's minds, get people energized and activated. Allow people to see their neighbors, their family members with more empathy in a different way. Without stories, who are we as a society? The way we tell those stories matters.
The intentions we have when we tell those stories matters, and who tells those stories matter. If I allows more people to tell their stories who haven't been having their stories told, then let's do it. But let's also remember that there's other aspects of how stories get told that is beyond the creative space. It's the distribution space. What work is getting distributed? What work is getting seen? It doesn't just happen in the editing room and on the film set.
That's so interesting. You know, a topic that we haven't actually touched on is bias in AI. Do you have thoughts, concerns, questions around sort of how AI models are being developed and with consideration of bias? I saw the film Coded Bias, which came out. I think it was it's been about 3 or 4 years now, so things are changing by the minute.
But it terrified me because it was based on the idea, a black woman working with. I had to put on a white mask in order for AI to see her. It literally was erasing her.
And so I think we need to understand that if AI is created by humans, it takes with it the same biases prejudices that humans put onto it. Yeah, it's a big topic of conversation for us internally is how do we ensure that the AI that we're creating, that we're developing is, is recognizing the pitfalls of biases? And how do we ensure that we're building models that are aware that we are building technology that is created in an ethical way? It's really important with content awareness. If you say, give me reaction shots, those people being chosen, reaction shots better be all the people versus action shots and better not be leaving out black and brown faces.
Absolutely. Stacey, thank you so much for being here today. It is so wonderful to have your voice as part of this conversation. And thank you all for watching.
It's important for us to continue to make space for these conversations, so stay tuned for more on AI in the video industry.
2024-10-01 14:55
