Vidpros Insiders
Vidpros Insiders take you behind the scenes of the creator economy.
In this podcast, we interview influencers, content strategists, social media managers, talent managers, and industry operators to give you an inside look at how the digital world really works. From growing an audience and building a personal brand to managing creators and scaling content teams, we break down the strategies, systems, and stories driving today’s content industry.
If you’re a creator, entrepreneur, or just curious about what happens behind viral videos, Vidpros Insider gives you the real conversations shaping the space.
New episodes weekly.
Vidpros Insiders
Andrew Ford Shares The Honest Truth About AI Video Editing in 2026
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Meet Andrew Ford, a software engineer, YouTube creator, and the creator behind Buttercut, an AI-assisted video editing tool built to help creators turn raw footage into rough cuts faster using Claude.
In this episode, we talk about Andrew’s journey from working as an engineer to taking time off to pursue YouTube, why editing became the hardest part of creating consistently, and how that frustration led him to build Buttercut.
We dive deep into AI video editing, Claude, creator workflows, YouTube strategy, storytelling, thumbnails, titles, AI slop, human taste, and what happens when editing becomes less of a technical bottleneck for creators.
Andrew shares how Buttercut works, why Claude alone cannot simply edit a video by itself, and how AI agents can transcribe footage, understand clips, summarize scenes, suggest narrative structures, and help creators build A-roll rough cuts without removing the human decision-making process.
We also discuss why editing is still about taste, why AI can help with mechanical work but still struggles with pacing and creative judgment, how creators can avoid becoming lazy with AI, and why the future of content may depend even more on strong premises, better stories, and being compelling on camera.
Andrew also opens up about his own viral YouTube experience, how he tested dozens of thumbnails and titles, why packaging matters, what creators misunderstand about AI editing, and why he believes tools like Buttercut could help more people make videos without burning out.
Whether you're a creator, editor, YouTuber, agency owner, filmmaker, marketer, or someone trying to understand how AI is changing video production, this episode is packed with thoughtful insights on editing, creativity, automation, storytelling, and the future of content creation.
Learn more about Andrew:
YouTube:
https://www.youtube.com/@maybe-andrew-ford
CONNECT WITH US
Website: https://vidpros.com
Instagram: / vidprosedits
Facebook: / vidprosedits
LinkedIn: / vidprosedits
MORE ON VIDPROS
Need pro video editing? Vidpros helps creators and brands produce retention-focused content that performs.
📩 sales@vidpros.com
#AIEditing
#ClaudeAI
#Buttercut
#AndrewFord
#YouTubeEditing
#VideoEditing
#ContentCreation
#CreatorEconomy
#YouTubeGrowth
#AIForCreators
#Storytelling
#VideoProduction
#AIContent
#CreatorTools
#VidprosInsiders
Andrew is a software engineer in San Francisco who spent a year trying to do YouTube full time and realized editing was killing him. So he built Buttercup, an AI tool that turns raw footage into a Roth Cut while you sleep. In this episode, he breaks down how AI is changing editing and why human taste still wins. The way I usually do this, I start just by asking the guests to introduce themselves. So just tell me a little bit about who you are, what you're doing today, what you're up to, what you're up to with Buttercut. And then from there we can start just talking. So if we could get started just for people discovering you for the first time and your work, how would you describe today what you're doing and what you're building in your career?
SPEAKER_00Uh yeah, my name is Andrew, and I'm a software engineer in San Francisco. Um, but I really, really like YouTube. Um and for like a year, I kind of took off time for being an engineer and just tried to do YouTube, YouTube full-time. Um, and the thing that I found is that it was so hard. Um yeah, it's just it's hard to get out videos regularly, it's hard to do it by yourself. Um, but I think the hardest part for me was just the editing process. Like um, I went to school for journalism, so I can sort of I can write a script and feel good about it. Um and then like, you know, at first being on camera is kind of weird, but once you get a few videos done, that kind of goes away. Um but the editing part never got any easier. Um it always took forever. Um I have one video that like, you know, I I sort of knew it was gonna do well, so I spent a lot of time on it, but I probably spent like five to six months on it. And if you want to do YouTube, that's just like not tenable. Um and I think also like, you know, my favorite YouTubers, you know, they just put out content a lot more often than that. Um so yeah, basically I'm full-time engineer, but side hustle is YouTube. Um and I just want to make it easier to do that.
SPEAKER_01Aaron Ross Powell What made you go into YouTube and actually AI-powered editing workflows and you know what makes you so interested in tools like Claude, for example?
SPEAKER_00Sure, sure. Uh well I think um they're separate. I think YouTube was just um I really like engineering work, but you don't have the same level of creativity that you do when you're making a YouTube video, um, especially as like a solo creator. You get to write the script, you get to make all the shots, like there's just you have all these like wonderful choices you get to make throughout the entire process. Um and with engineering, you're often kind of just getting like assigned tasks from somebody else that you've got to do. I also, since I had like a background in journalism, like I just enjoy storytelling. And I think that like YouTube is the best medium for that right now, um, you know, short of having like lots and lots of money to do it. And yeah, I think the reason that I got kind of got into like Claude and editing that way is just to speed up the process a lot. Um yeah, I kind of feel like when I took that year off to try and just do YouTube full time, um overwhelmingly the time spent was on editing the footage. Um so I just wanted to speed that up to make, you know, I think there was this time when when Casey, Casey Neistat was putting out daily vlogs, um, and then a lot of other people were like, oh, okay, let's let's try this. Um and I think they were successful for a while, but they just got burnt out because um, you know, one, it's hard to be on camera every day, but then also like the editing part is just um it's slow and it takes a long time. So so I sort of want to like bring back the ability to shoot vlogs and not have them take forever. Um I was also sort of inspired by uh I think it's Chris Meets Chris. He's this um YouTuber in Canada, and he does these like almost A-roll style vlogs where it's just like jump cuts, A-roll with a GoPro, um, but it's still pretty entertaining. So I was like, oh, I think, you know, if you got it down to just being A-roll, like you could automate that now with uh with Claude.
SPEAKER_01How important do you think editing is? Do you think a video should be like extra edited and overly edited and all crazy? Or do you think we can do something very minimal, very simple if we're focusing on quantity instead of you know, quality? Weird to say it like that.
SPEAKER_00Yeah, I mean, yeah, you know, I I find uh this one YouTuber, Chris, I find his interesting, even though they're mostly A-roll and there's not that much editing. So I think for sure, like, you know, I think Casey set a bar that was too high for anybody to maintain, including himself. Like he also burned himself out. Ultimately, it kind of comes down to like the story you're telling and the person that you're following around. And if you like their story and like them, then you'll kind of watch them even if the editing isn't perfect. Um, with that said, if you want a video to, you know, go viral-ish, um, you're still gonna have to put in the time to make um to kind of keep the pacing up and keep retention high. Um so you know, you know, a little bit of both.
SPEAKER_01What do you think takes for a video to go viral? Because you've had that experience one with a video that did really well. What was that that experience like for you? Did you have a strategy going on about it, or were you just lucky, would you say?
SPEAKER_00No, definitely not lucky. Um, I definitely went in with a strategy, which was just um like the most important thing is starting out with like a compelling story. Um, and for me, the story was basically I was doing this bike YouTube channel, um, and I had bought like a fancy e-bike for like five grand, and we had a fair amount of bike theft in San Francisco. And what I hated about it was that this fancy bike didn't come with the GPS on it, so that if a bike theft took it, um, I could get it back. So basically the premise of my video was if a bike gets stolen with GPS, can we get the bike back? Um, so I already knew that was kind of like a juicy idea. Um, and I had seen other videos, you know, do similar things. But yeah, so the most important thing is just starting out with like a compelling story and premise. Um and so I went into it knowing this was kind of like an interesting idea and it would just kind of come down to um like execution if I could pull it off. I didn't know like how many views it would get, um, but I knew it would, you know, perform reasonably well.
SPEAKER_01When you were about to post the video, did you think about SEO a lot or were you just coming up with titles?
SPEAKER_00I don't know if I was thinking I mean, yeah, I I was thinking about thumbnails and titles for sure. Um because I did make uh, you know, the video now has something like 600,000 views, but the first three months, I bet it had 500, a thousand. So it definitely took a while, and I um I tried something like 15 to 20 different thumbnails um before it finally kind of like took off. Um I don't know if that was like if I would have just kept one of the worst thumbnails, if it also would have worked out. Um but I was definitely like churning on thumbnails and titles. So that was kind of like my only SEO optimization for it. But titles and thumbnails, that that super matters because that's like the preview of the premise of the video.
SPEAKER_01Can you walk me a little bit into what it's like editing with Claude? How helpful is it? How how how do you do you do it? If people are watching this and they're like, I might give this a try.
SPEAKER_00Right now there's like a few different there's basically maybe two, three steps to it. So the first step is that um Claude alone cannot edit video. Um like maybe if you had infinite time and an infinite subscription, it could figure something out. Um but if you just give Claude your video footage, like it's not going to come up with an edit. So basically what I built is Buttercut, which is it's basically just a folder that you open up in Claude, and then it gives it like a structured way to think about um how to process the footage and then edit the footage. Um so the first step is that you just give it like a folder of footage, and then it basically first transcribes all the video and then it watches all the video and then it summarizes all the video so that um it can basically have like all of the footage in its head at once. And then once you have the library processed, then you just kind of talk to it about like what type of um timeline you want to build. So I think like the the one that's the most helpful for me is just building out that A-roll track because like there's been so many times where I shot a video and then like a month later I can't really remember what's in the footage or like what exact my narrative was, or I during the day I had this like you know, perfect script, and then when I actually shot it, you know, I I couldn't do it because the locations were in construction or whatever. So just like while you're shooting and it changes. So anyway, the nicest thing is to just give it all of your footage, it looks through it all, and then you can say, build me an A-roll rough cut. And then it will like find the narrative in there or multiple narratives and say, like, hey, would you like me to structure it like this? Would you like me to structure it like this? Um it basically gives you like three options, and if you'd like one of those, you can just move forward with it. Um, or you can say, like, oh no, it actually should be structured this way. Um but basically it's just it processes your footage and then you just talk um talk through it with it.
SPEAKER_01In the process, was there anything that you felt like surprised you when it comes to AI and AI editing in general?
SPEAKER_00Yeah, I think what so I kind of have this like one vlog that I shot that I'm just constantly making more rough cuts on. Um and I think like the first time it kind of surprised me was that I shot it as just like a linear narrative, like I start in my neighborhood, I tell you some backstory. Um then I shot like some b-roll that I thought would be like a little montage. And then I get to this like meetup. Um and then at the end I have like a summary. And like I kind of thought, like, oh, that's the only way to edit it. Um but one of like the first times I was talking with it, it basically was like, oh, we could shoot, we could edit this linearly, or we could even just start with the end of the story. Because you kind of like start with the end, the um, the payoff, and then go forward. And uh it was just really nice having like a sort of assistant editor um sort of take you out of that like tunnel vision of like this is how it must be edited, and just give me like another way of of working with the footage.
SPEAKER_01And is there anything that you feel like AI editing agents do well already that most people still underestimate? Because you know, people are scared of AI. So I do want to know your take as uh on this, and also I want to know as a journalist yourself, how do you feel about this and AI? Because people are scared, so I usually ask for the guests to let me know how they feel about this and what they think the future holds with this.
SPEAKER_00Yeah, I think the thing that they are really good at is they can absorb um just so much footage at once. Like, you know, I feel like traditionally when I'm editing a video, um, usually it's like at night after work and I'm already tired. And, you know, to get like one 30-second scene out, I feel like I've got to scrub through all the footage and just like remember what's there. Like there's definitely some magic in that because you'll s you'll connect shots that um you know you didn't initially intend. But like when you're editing yourself and you haven't edited in a couple days, it just takes a while just to like get re um what's the word? Like re-initiated with the footage. It's gotta be something better.
SPEAKER_01Restarted.
SPEAKER_00Yeah. Okay. It takes you a while to like digest the footage. But AI can just like read every single transcript and look at every single timeline like all at once. Um so it can kind of just like almost instantly grok it. Uh, let's not use grok. It can instantly um understand it in a way that's uh it's it's a lot harder for us.
SPEAKER_01Aaron Ross Powell And what do you think they're terrible at that humans can do better?
SPEAKER_00We're still just better at coming up with the premises of stories. We're also our taste is so much better than AI. Um I've kind of tested Buttercut with like Claude, Codex, and Gemini. And Claude definitely has the best taste, um, but it's still just way worse than humans. I think like one interesting thing that I don't know if everybody knows this, but like uh, you know, programming and engineering is uh everybody is using you know Claude or Codex all the time to sort of automate their engineering. And if you use some feature on a website that's a new button that goes somewhere, like you don't know the difference between like human-generated code and uh claude generated code. But if you watch a video and you have like decent taste, you're going to immediately be able to tell that just like something it just doesn't have the right um like pacing and taste and feeling that humans do. So it's it's great for getting the initial narrative out, but like we still absolutely need taste to get a decent video out right now. And I think it's gonna be that way um at least for the next couple years.
SPEAKER_01And how would we justify the fear that we have in society right now when it comes to AI and AI content? And as a journalist yourself, how do you feel about this?
SPEAKER_00Uh I think it's it's all totally uh justified. Uh it's it's very scary for me. It's um uh it's crazy how different it's made my career in just you know a year, year and a half. Like it's gone from being like barely helpful to now able to just automate so much of my career. Um so I think I'm like kind of on the front end of that being a programmer. But I think it's gonna come for most desk jobs. So yeah, I I think it's fine to be scared about it. I think it's fair to be justified. If you don't want to use any AI, uh I wouldn't judge you at all. Um but I want to bring back like making vlogs uh sustainable. And like if you've got to, you know, spend 40 hours to edit a six-minute video, um, it's just like not that practical.
SPEAKER_01And do you think there's anything that people misunderstand about AI video editing in the specific?
SPEAKER_00Yeah, yeah. I definitely get nasty comments. Um I think that people don't realize that you can still like take its work and then like apply your own taste to it. Um I also think that like some of the best movies of all time are like very similar from the script to which to what the actual movie is. Like some directors, you know, they shoot tons of coverage and you know all this, but I think like a lot of the best directors, like their script and the output is very similar. So if you write a script before you make a vlog or you at least have an outline, like you can still give that to Claude, give that to Buttercut, and say, make a rough cut with this. And it's really just doing like mechanical work for you. Um, you're still gonna have to have the taste to clean it up and make the pacing nice. Um, but like this is not handed off to the AI and then you get like some viral video. Like right now, it's really just like automating the grunt work of making that initial A role, maybe finding B roll for you, and then also like getting you sort of like into the zone of the edit quite a bit faster.
SPEAKER_01When it comes to, like you said, using Claude to edit movies, what did that process teach you like about how editing actually works? Did you feel like you learned something with just by watching AI doing its job?
SPEAKER_00Yeah, I mean, I think I more just see where it's bad. Um so I think like, you know, something that comes, I think, naturally to somebody that um has watched a lot of videos, which is everybody, and then tries to make their own edit, is there's like a natural cadence to pacing that feels nice. You know, there's like whatever that rule is where it's like you want to look for three seconds, you want to blink, you want to see a different shot. Um, and that kind of like naturally comes to humans, but that does not come naturally to uh AI or Claude right now. Um that isn't to say that I don't think we can sort of replicate that by just giving it instructions on pacing and stuff. Um, but naturally it does not have that in the same way that we do.
SPEAKER_01Uh do you think it's scary for video editors, let's say, when it comes to AI now? Or do you think there's always a space for video editors?
SPEAKER_00It's weird because I think of video quite a bit different than I think of engineering. I think of engineering, uh programming, it's gonna get automated. Um, all the programmers are very scared, the job market is bad, um, every company has their engineers are getting so productive that they're just doing mass layoffs. With video, on the other hand, I'm I'm hoping and I have like a hunch that it could be the opposite, which is to say like there's lots of creators out there that you like as people, you like their story, you want to hear their stories that you don't know about, um, but just like they don't have the time to edit. Um, so I'm hoping this is gonna like bring out more creators. Like when I'm thinking about my tool, I'm not thinking about like, oh, how can we make uh generative AI robot videos? I'm thinking like, how can we take like real footage um and get like a rough cutout faster? So I think like unlike engineering, I think this could actually be like a boon. Is that the right word? This could be like good for both editors and video in general. Um yeah, I I have almost every like fancy, you know, streaming subscription, and still almost every night I get on YouTube. Um so yeah, I don't know. I feel pretty good about editing.
SPEAKER_01Do you think editing is more about taste or technical skill? What do you think should come first?
SPEAKER_00It's definitely a hundred percent about taste. Yeah, because uh you know when you make some of your earliest videos, you're gonna play them back and you're gonna be like you're gonna be ashamed of them. Um and that's because your taste is good, but you just don't have the technical ability um or like the processes to make to make the video good. Like maybe just the footage itself is bad, the audio is bad, whatever. But I think um yeah, it's it's really taste dependent.
SPEAKER_01Yeah, I I one thing that I always say that I always I always say this about AI and also the internet in general. I feel like especially for filmmakers, it made everything much more accessible, approachable. Anyone can try doing this and anyone can be a part of this. So, where do you think we're moving forward right now? Do you think um we're going towards a direction that creators can operate like a full media company? You know, people with you know less opportunity. Do you think they could actually build something massive using those tools right now?
SPEAKER_00I mean, if you just think about it as, you know, you're working, you know, a 50-hour a week job to make ends meet. Um, so all you've got is, you know, Saturday, Sunday to shoot and edit. Um before that was like if you were a hardcore creator, you could make it work. Casey could make it work. Um, some of the hardcores can make it work. But I think what's going to happen here is that, yeah, you'll you'll have way more leverage. Um so that yeah, with with the same amount of time, you'll either make a better video, or I think more likely, like people that just couldn't get the videos out before will now hopefully be able to get those videos out. I mean, I think about like all these like weird niches, like tiny homes and crocheting and uh, you know, cabin videos. Like, um, I bet there are people that are even better at making uh those stories, but they just don't have the time to edit. Um, so if we can get it so they can write the script in an hour, shoot in an hour, and have a nice edit in an hour, like that would be that would be a big win for, you know, at least niche YouTube.
SPEAKER_01Do you think there will still be some sort of bottleneck? And where will it be if we actually solve the editing process and remove that?
SPEAKER_00I think the only remaining bottleneck would be just there's only so much of us, there's only so many people on Earth, and we only have so much time to watch YouTube.
SPEAKER_01Yeah.
SPEAKER_00Like we only have so much time for entertainment. So we have like we can watch Game of Thrones, um, or we can learn how to edit in Final Cut. Um, and we kind of always have to make those decisions. If I was to guess, like the bottleneck would be coming up with good stories and good scripts, that I think will be the bottleneck. Because I think that's also the bottleneck in Hollywood. Like there's there's plenty of very good actors. Um, and I know this because I see good actors in bad movies all the time, and you're like, oh, that's a that's a bad movie, that's a bad actor. But then you see them in a good movie and you're like, oh, it was just we didn't have a good script. Um so yeah, I think the two remaining bottlenecks would be how much can people watch YouTube, and then how many good stories are there to tell.
SPEAKER_01If you're a content creator and editing is slowing you down, listen up. VidPros is a professional video editing service built specifically for creators, not random freelancers and not AI tools. You get a dedicated human editor who learns your style, pacing, and brand from day one. You film your content and upload the raw footage, your editor handles everything cuts, captions. Color grading, sound design, and short form clips. We deliver everything through Google Drive and frame.io so you can leave time-stamped feedback and requests and limited revisions. We start with a $100 trial week that includes 10 hours of editing. Go to vidpros.com and start your trial today. Stop editing and start growing. Yeah, because there's also a lot of AI content out there right now, and we all know that the quality is not the same usually. There's a lot of AI slop. So how do you actually feel when you watch those videos that are very much generic AI content and in a way emotionally empty? And how do you, with your strategies, solve that?
SPEAKER_00Yeah, yeah. So this is interesting to me. Over the past few days, um, you know, I watch a fair amount of camera videos, and there's a ton of camera gear videos that are so clearly um AI script, AI voiceover, AI B-roll. And right now, every single time one of those channels comes into my home feed, um, I just block the channel. So like I, the like low effort stuff, I don't know if everybody, I know that everybody is in this way, but at least for me, like I just block the channel entirely. Um like I just, you know, I've only got you know an hour or two to watch YouTube. Like I want to watch, you know, high quality stuff that people um you know put sweat and tears into. So I hide them. But yesterday I was looking at one. Um, you know, it was a video about the new uh the new Pocket 4 Pro. Do you know about that camera?
SPEAKER_01No, I don't.
SPEAKER_00Okay. Do you know the DJI Pocket? It's like the Yes, yes. So anyway, they came out with the four, it's good, and then they're coming out with the Pocket 4 Pro, you know, in a few days. But I saw so many comments on the video that looked like people were just like, oh yeah, that's just a camera review video, and they're just reviewing it. And I'm like, how uh like how did you not immediately smell this out and stop watching? Um But anyway, yeah, I don't watch them. I've only got an hour or two, I want to watch it from from real people making real stuff.
SPEAKER_01Aaron Powell Do you think it's a prompting issue, or do you think it's there's a deeper limitation with that?
SPEAKER_00Aaron Ross Powell There's a deeper limitation. An AI, like if they're writing a script for something, it's gonna be like there's two problems, but it's gonna be the script is gonna be average because like they've read the entire internet, they've read every novel, um, they've read every Reddit post. So when they write a script, like it just it feels very average or it feels very marketing-y. You can sort of guide that, uh, but it's just we're more interesting because we've, you know, had our trials and tribulations and gone through our lives, you know, with all these experiences that make up how we see the world, and that is more interesting than seeing all of the world at once, um, you know, like the AI has. Like, I just we're more interesting.
SPEAKER_01Do you think we're gonna get to a point that we can actually teach storytelling to AI so they can do it better? Or do you think we can already do that?
SPEAKER_00It'll probably be able to get there, but like it's gonna require it to be able to like go out into the world and like see stuff. So that will be hard. Um, like that's I don't know how long it is until we have robots walking around with um DJI pockets, like shooting vlogs, but I think we're still a ways out. Um I also just think like uh you know, both as a watcher of YouTube and as a creator on YouTube, like you have empathy for the people you're watching and for yourself when you're shooting it. And it's gonna be harder for AI to just like fake empathy in general. And I think like empathy is sort of like again what makes interesting stories.
SPEAKER_01Yeah, this can become some sort of philosophical question if you think about it. Like, is taste inherently human? Or, you know, can AI ever approach things the way that we can? So, yeah, that's something interesting that I don't think anyone has an answer yet, and we're still figuring everything, everything out uh along the way. And I do think the fear comes from that too, of like we don't really know what's happening. Like, I usually ask guests, and it's a question that it's always hard for everyone to answer. But like for you, in five years from now, where do you think we're gonna be with AI? If you were to guess.
SPEAKER_00Five years is too far out. Yeah, man. So I, you know, the I don't know, this would get like to programmer world, but like, you know, the consensus in programmer world is there's gonna be like artificial general intelligence in like 2027 or 2028. Once they get that, then who the heck knows? I, you know, my only dream is that we all get some universal basic income. Uh, you know, we all get 70k a year. Um, I've got a robot to make me French toast in the morning. Um, and then I just go shoot my vlogs, um, come back home, give it the raw footage. It gives me, you know, three different rough cuts. It gives me a you know traditional narrative, it gives me a circular narrative, um, and then it gives me a you know uh heavy voiceover uh narrative. Um and then it makes me another awesome meal for lunch. Um and then I, you know, pick up the kids from school. Um if if we get to there, that would be fine. But is that how it's gonna play out? Like, are the governments gonna pay us to just make vlogs? I don't know. Um, I don't think I'm gonna be programming in five years. Um I think I'll still be doing YouTube though. What's your fear? What do you what do you think in five years?
SPEAKER_01AI in five years. Oh, nobody asked me that question, and I've asked this question a lot. Um I do think that we're gonna have you know a specific genre of content when it comes to AI content. For example, I do think we're gonna have some big movies made entirely with AI. And I'm talking about actual movies running for the Oscars and festivals, and it's gonna be controversial, and but I do think it's gonna be more accessible. But I I I don't see a way of AI not entering mainstream media and staying, because I do think AI came into this world to shift things, and even though a lot of people might be against it, and I understand the reasons why, I don't know how we cannot work side by side with it and compete with it now.
SPEAKER_00Can we go back to your movie? Can I ask a couple follow-ups? Okay. But we're talking about in five years' time. Okay, five years' time. The thing is in the next two years, I think AI could get super great at programming and building bridges and doing math and physics, but I still think it might have bad taste. In five years, I agree with you that it might have good taste.
SPEAKER_01Yeah.
SPEAKER_00So yeah, five years is too far out. Five years, yeah, I wouldn't be surprised if if there's um AI scripts that are quite good. I still feel like that's a limiting factor. Like now you can generate video well, but nobody's making better scripts.
SPEAKER_01Yeah. No, if I'm being honest, I would be against an AI movie at the Oscars, for example. I understand why, but I do see how things are shaping for AI and the direction that we're going. And I think a lot of the things that are happening are inevitable with AI, and we just need to adapt and work with it. But um yeah, I I do do you think when it comes, for example, to editing, do you think AI actually like they understand about pacing and retention in a way they might not understand about the the structure and the script, but do you think they get something right when it comes to that?
SPEAKER_00So by themselves, no. Uh right now, especially if it's just a roll, like it will just give you the whole sentence. However, if you merely tell it, like, here's what natural pacing is. Um, you know, you begin a sentence with uh three seconds on the person, then you switch to a B-roll clip that's you know, um a second and a half, two seconds, then you switch back to A roll, then you show another B-roll shot for a second and a half, then another B-roll shot for a second and a half, like it will absolutely follow those rules. So even though it doesn't sort of inherently know what feels good, um you can definitely tell it what to do, um, and it'll follow your rules. So I think like one of the things on the horizon for Buttercut is just um, you know, depending on like what type of video somebody is editing, giving good guidelines for pacing to butter to Buttercut, so then it edits it well. Like that doesn't help you magically come up with more B-roll. Like you've still got to have sufficient B-roll to provide that pacing. Um but you can definitely tell it pacing and it will follow that. So it almost feels like it knows what pacing is. You just have to tell it once and kind of have it stored in its memory.
SPEAKER_01One thing that I want to know too is do you think packaging matters even more now that content production is getting easier?
SPEAKER_00Um I don't know if it matters more, but it's just always been essential. Yeah, you know, so you know, in college when I was doing um journalism and like laying out newspaper pages or you know, blogs on the internet, um, like we thought about headlines all the time. That was kind of like the image and the the headline, that was kind of the packaging for newspapers. Um and for blog posts, it's the same. Uh for YouTube, it's no different. You've got to have a compelling title and thumbnail. Like, I don't know when you say packaging, do you just mean those two things? Or do you think even more broadly than that? What else am I missing?
SPEAKER_01I would say more broadly.
SPEAKER_00What other kind of things would you include in there?
SPEAKER_01It's really hard bringing this question back to me because I'm really good at asking the questions, but I don't think about that world that much. But one thing that I'm thinking right now, if I were to think, I think if we go back, packaging is gonna be way more important too, I think, in a way. And the reason why I think that is because don't you think that with production getting easier, it's gonna get a little bit harder to earn attention and we need to find other aspects that we uh that we can focus on to earn that attention. What what do you think?
SPEAKER_00Yeah, if the quality goes up, like it's always harder to it's harder to get eyeballs, but I still feel like you know, I follow, I don't know, let's say 50 creators.
SPEAKER_01Yeah.
SPEAKER_00And when I go to just my subscriptions feed on YouTube, um still most of their videos I don't want to watch. Um so I think like if you make a good video, like the opportunity is absolutely still there. Um but yeah, like the just to make it like really literal, like the premise of the video matters a huge amount. That's like the most important thing. And then the thumbnail is sort of like how do you tease that premise? Because you can change your thumbnail without changing the premise of the video. And then the title and thumbnail like obviously play hand in hand too. Because like you can sort of do like an information dense, like long title, um, or you can do like a short one and then have like the thumbnail sort of be the second part of that. Um, but just to make it like ultra tangible, yeah, having like a great premise and a great thumbnail and a great title, like it you your videos will not get any views without it unless people are like really dying for it.
SPEAKER_01Do you think it should be shorter or longer, or do you think it's gonna depend? How do you earn that?
SPEAKER_00You know, I prefer the title to be as short as possible. You know, I don't know what like YouTube class I took, but it's like, you know, it made me think about how I use YouTube, and when I go on YouTube, you know, I'm scanning and I'm not spending that much time. Like, so I don't know exactly what my word count is, but like, yeah, the shortest, the shorter the better, and the clearer you can communicate that premise, the better. Um, so use active voice. Um, yeah, the shorter the better.
SPEAKER_01One thing that I want to know, let's say you're running a production company uh or a video editing agency, how would you adapt for the AI era?
SPEAKER_00Depending on how long your A-rolls take to edit, like that's a great start. The good thing is that like, you know, professional editors, like I assume you're a professional editor, like you edit quite a bit. Um, you're fast at getting out an A-roll, but I think people like me that are sort of just like amateurs and are perfectionists and are editing videos about themselves, um it's hard. So I think like if you're an agency, like you just want to do, you want to leave taste to your editors, but you want to like remove the mechanical part. So if you do have like a script that you're getting from your um from your client, like you can absolutely get the script to the A roll, like you can have that automated right now, and it'll be, I would guess, faster than your your editor could do it. Um I think the other thing too is it's just it's really helpful for brainstorming. Like you've got, you know, however many minutes or hours of footage. Um, and as an editor, you've got to like figure out how to structure the story. Instead, you can talk to AI and be like, hey, what options do I have to structure this? You can ask it, what is the best? Um you're gonna be able to make a better decision on that, but you can at least tell it like, how can we structure this video if the client isn't like this is how it must be edited, which I would think most clients that are handing over their footage, they're not gonna be as sticklers for it being exactly their vision. They're more, they're putting more of a premium on volume than they are on it being like exactly frame by frame how they want it to feel.
SPEAKER_01Yeah, one thing that I also think about is you know the turnaround time. Will clients expect AI speed turnaround time when it comes to their projects? And that might be the case. So uh but one thing that I find interesting, even the way you edit yourself, of course, we use AI, but it's some sort of assistant, it's a tool. It's not about just like getting AI to do all the work for you. Of course, that's not gonna work. So yeah, AI can do some sort of foundation, but we can still like refine things, and that's what you do today. And how do you avoid creators becoming you know creative, creatively like I would say like a lazy creativity type of scenario when it comes to AI? How do we avoid being lazy and actually creating a content that you know performs well, that is good, and use AI to help us, if it made any sense what I just said.
SPEAKER_00Yeah, yeah, sure. I mean, I I don't think you can force people to get lazy with it, uh or you can prevent them from being lazy. Like they can be lazy, um, but that's just gonna show up in the edit. Um I think like the way to avoid being lazy is to like have you know big dreams and big goals and to have good taste. Um and I don't think good taste is like intrinsic. It's kind of like um exposing yourself to, you know, creators and directors and writers that you like. Um basically just like surrounding yourself, like not in person, but like surrounding yourself with art that you like and that you sort of like aspire to get to. Yeah, one of my favorite things to do um if I'm not inspired, is uh to like go to an art museum. And even though like I'm not a painter, I don't draw, I don't do anything like that, like I always like I see colors and I see themes and I see like textures that like I want to implement into like my own work. Um so yeah, the way to not get lazy is to have you know good taste, expand your taste, um, and then to have sort of like big dreams of like getting somewhere with your work.
SPEAKER_01Yeah. So being inspired, right, in a way, that's the secret to not being lazy. Do you think audience audiences will actually eventually recognize AI generated editing patterns? Or do you think when it comes to editing that's more difficult?
SPEAKER_00I mean the thing is, is like there's there's only good editing patterns, I think. I don't think there's necessarily AI editing patterns. Um So I don't think you'll be able to tell. I think you'll you just be able to tell that the pacing or the taste is off. Just because like it's gonna be so easy in the future to just tell AI, like, you know, make this the pacing of Casey, make this the pacing of um a documentary, make this the pacing of The Godfather. Um that's not gonna give you the script, that's not gonna give you the footage, but you're definitely gonna be able to tell it like the pacing that you want. Um, but there's still like a taste and decisions on things. Like I think every editor has had a time when they like look at a clip and it's 25 frames and they look at it and they're like, oh, it's two frames too long. Like I need to just clip it back, these two frames. Um and AI is not gonna clip it back those two frames for you right now. Like you're still gonna have to see what's wrong and correct that yourself.
SPEAKER_01Do you have like any influencers, creators, companies that you think they're actually adapting to AI like in an intelligent way right now?
SPEAKER_00No, I don't. Um I I yeah, I I I think like when a AI is done right, like you don't you don't notice it. So maybe behind the scenes, they're spending less time on their edit, or maybe behind the scenes, like they came out with a better movie because they had like a more interesting premise or a better script. But if you do this right, it's a tool, um, and it will, you know, either speed up the process or improve the quality. Um, but we as like the viewers shouldn't, it should not feel AI at all to us. It should feel, you know, like just a more polished human that got to make French toast the next morning instead of like staying up all night uh you know to finish the edit themselves.
SPEAKER_01Is there any belief that you have about the future of creators or editors or anyone in this industry but most people would disagree with?
SPEAKER_00Yeah, I think um I watched one of your interviews. Which one? I can't remember, it's like a husband and wife guy. Uh I anyway, he he was just saying like he was saying like the old rules don't work. You can't do a video a week and expect that to work anymore. And I would disagree with that. I still think if you come out with a great video a week, like that's more than enough for me to subscribe and watch every single one of your videos. If if Casey just put out one video a week instead of disappearing for six months, um, we'd still watch that video a week and I think he'd still be rich based off of YouTube. So I actually like don't think that, you know, just because AI is here, I don't think we need to change our flow to put out, you know, daily videos now. Like we can't just do one video a week anymore. Um I don't think we need to like have 10 shorts out a day. Um I think we should just aim to like make the videos we're already making better. Like it's just so much better to have a video that gets, you know, a half million views that's nice and polished and you know, you're crying at the end um than it is to put out like five bad videos that get 5,000 views. Um yeah, that's my only thing. I just because the tools are there doesn't mean we've got to get like overwhelmed and and think now we've got to be productivity monsters. Like ultimately we're doing the oldest thing, maybe not the oldest thing, but we're doing something that's been around for like a hundred years.
SPEAKER_01I I have one final question, actually, because uh and this goes like in a way, kind of like tying up everything that we talked about and about the future and this uncertainty. But if AI eventually handles execution and everything about it, what should creators and editors focus on becoming exceptional at? So where do you think we can actually go from this? Where do we show our individual value in a way?
SPEAKER_00Yeah, totally. I mean, the the nice thing about video is that it's still like a three-step process. You've got like the script, you've got shooting it, and you've got editing. Like if we clear out editing and it just gets totally automated and it, you know, AI figures out the pacing and cuts back two frames and gets the audio's levels right, like we've still got two whole things to work on. We've still got the script and we've still got us on camera. And there is so much to explore in that. And I don't think anybody that's ever shot a YouTube vlog has ever been like, that was it. That was the perfect vlog. The pacing was perfect, the shots were perfect. Like there's so much to do with making our scripts better, making the stories better, and making uh and just shooting better. You know, not to mention like how many of us have like a notion document with like 20 video ideas that we can't do. Um so if we automate editing, which we're not gonna have automated for a couple years, um, like we've still got a ton of uh creativity left to ring out. Yeah.
SPEAKER_01Awesome. Honestly, Andrew, thank you so much for doing this, saying yes to doing this. And if there's anything you'd like to promote that you're doing right now, just please let us know. This is your moment.
SPEAKER_00Right now, Buttercut is free and open source. Um, I've been working on this since December. It's free right now, but it's not gonna be for forever. I'm probably gonna price it like a LUT, you know, maybe like 30 bucks. And then hopefully, you know, enough people buy it that then I work up to maybe a hundred bucks. Um, you buy it once, but for right now it's totally free. You can just download it, open it up with Claude, process your footage, um, get a nice little A roll, and then I'm working on this all the time, so it kind of gets better every few days. Um yeah, buttercut.io and join the Discord too.
SPEAKER_01Awesome. I'll actually leave the links down below and thank you again for saying yes to doing this.
SPEAKER_00Cool. Thanks for your time, appreciate it. Nice to meet you.
SPEAKER_01The script and the shoot are still yours. AI can handle the grunt work, but the story only exists because you lived it. Subscribe, leave a comment, and I'll see you in the next episode. Bye.