On this episode of the AI For All Podcast, Zohar Dayan, SVP of Product at Vimeo, joins Ryan Chacon and Neil Sahota to discuss AI in video and the future of content creation. They talk about the history and evolution of AI in video, what videos AI is best for, the demand for video content, automating video creation, the Hollywood strikes, making video accessible to businesses, standing out in a world of AI-generated video, and tailoring video content to different audiences with AI.

About Zohar Dayan

Zohar Dayan is head of Vimeo’s video creation and editing products – enabling the world’s leading companies to unlock the power of video for effective marketing and employee communication. He is also the CEO and founder of Wibbitz, the leading automated video creation company that empowers any team to produce professional short-form video, which was acquired by Vimeo in November of 2021. Wibbitz Studio powers video for leading brands including Verizon, Bloomberg, Hubspot, Conde Nast, and S&P Global. Dayan launched several startups at a young age and now has over 20 years of experience in product design and management. Dayan has spoken at SXSW, VivaTech, Mobile World Congress, the World News Media Congress, and European Newspaper Congress, and gave a TEDx talk on breaking old habits to inspire innovation. He was previously selected by Geektime as one of the “Top 5 Israeli Entrepreneurs” and the 40-under-40 list by TheMarker, Israel's leading financial magazine.

Interested in connecting with Zohar? Reach out on LinkedIn!

About Vimeo

Vimeo is the world’s most innovative video experience platform, enabling anyone to create high-quality video experiences to connect better and bring ideas to life. With a global community of nearly 300 million users, Vimeo empowers everyone — from creative storytellers to globally distributed teams at the world’s largest companies— with the ability to create interactive, video experiences.

Key Questions and Topics from This Episode:

(00:37) Introduction to Zohar Dayan and Vimeo
(00:49) AI in video and history of AI video
(02:30) What type of video is AI best for?
(05:17) Demand for high quality video content
(08:52) Automating video creation
(15:22) Hollywood strikes
(16:33) Leveling the playing field also raises the bar
(18:38) Standing out in a world of AI generated video
(20:23) Tailoring video content with AI
(21:56) Learn more about Vimeo and its AI tools

Transcript

- [Ryan] Welcome everybody to another episode of the AI For All Podcast. I'm Ryan Chacon, and with me is my cohost, Neil Sahota, the AI Advisor to the UN and founder of AI for Good. Neil, how's it going?

- [Neil] Doing all right. How's everybody else doing out there in podcast land?

- [Ryan] And we also have Nikolai, our producer, with us to chime in with some questions, especially on this topic, which is a video conversation.

Today's episode, we're going to be talking a lot about AI in video creation, how businesses can benefit from it. Just the history of AI moving into the video space. Talking about generative video. Lots of very interesting topics that I think our audience is going to get a ton of value out of. And to discuss this, we have Zohar Dayan, the Senior Vice President of Product at Vimeo.

They are a video experience platform, but I'm sure most of you out there already know who they are. Zohar, welcome to the podcast.

- [Zohar] Thank you very much. It's great to be here with you guys.

- [Ryan] Yeah, it's great to have you. So to start things off, what I wanted to do is ask if you could give our audience kind of a history of just AI in video creation. Like when did AI start to get involved in video?

Where has it evolved into now? Just to set the stage for our discussion.

- [Zohar] Yeah, so I actually, before I was at Vimeo, I was the founder and CEO of a company called Wibbitz. And we essentially created an automated video creation and editing platform. This was back in 2009 when we started the company.

I think it's very relevant to the conversation that we're having today and we're seeing around generative AI and video. But these attempts to automate video production and content production in general is not new. Back then we actually I think we were the only company that we coined the term text to video which we see today is pretty used pretty frequently. Text to image, text to video with all the prompt based generative AI stuff.

But I think that the notion of what automated video creation using AI means today, I think it started more of automating a lot of the process of video editing and video creation. Where today, I think it's more in the context of really generating something out of nothing.

So I think in that sense it's different, but I think the underlying notion is very similar. Creators want to be more productive. They want to be more efficient. Businesses want to be more efficient in how they approach content creation, and I think this is the next step of that evolution.

It started with tools, kind of lowering the barriers to entry for creation and evolving into automation and what automation means.

- [Ryan] So are there types of videos that the what we're talking about is lending itself more to being created or are there like where the limitations and where's the main adoption that you're seeing for AI and video?

- [Zohar] I think it really depends. There's really a wide array of tools and solutions and ways to utilize AI when it comes to video. You have the hot stuff that is very trendy right now with the generative AI which I think is very much in the early days but progressing very fast.

And I think that would influence a lot of that creative world. And I think any type of market segment or industry will benefit for that as that evolves and becomes better. I don't think that the material that is produced today is quite at the level to be production ready and that will see like a major brand using it in their next commercial or the next feature film using it.

But it's definitely approaching that. And I think it will get there sooner than we think. But when you think about AI when it comes to creation and editing, there's a variety of different use cases and tools out there. Some that we provide through our One Take video creation suite, which is more around how do you get started faster with video creation?

How do you automate a lot of those processes? How do you leverage AI to fine tune, tweak, clean up, and enhance your videos in a much more efficient way? And by that also again lowering the barrier to entry to video creation and video editing. Historically, when we're talking about video editing, if you didn't take film school in college or pretty much dedicated your life to learning these advanced video editing tools like Final Cut or Premiere or After Effects or DaVinci Resolve, like it's pretty much impossible to get started with that. And I think a lot of these tools that we're seeing being introduced to the market really level the playing field and democratizes video creation and video editing for everybody and for different functions within organizations.

There's more use cases for videos within an organization. I think the expectation of what is required or expected skillset of employees across the board is changing as well. And obviously, video was very much and it is very much a marketing tool for content marketing, for social media, for stuff like that. But it's really expanding into many other areas like executive communication and thought leadership. Even the way we run our meetings and events and the way we communicate within organizations today is primarily video.

- [Neil] I think you've created such great tools.

Like I appreciate you guys have the ability to strip my ums and ahs out, but I also think it's raised the bar now in terms of quality, right? You talk about executive communication, social media posts. People want the high quality video and the whole nine yards. I think it's great you guys provide a lot of tools around that.

Are you seeing from like your customer base, client base, the increased demand around like video, quality content.

- [Zohar] 100%. I think the more people can create more polished video and more professional quality looking video, I think the more people want to level up to that.

And I think it's becoming an expectation. And I think, going to my point earlier, I think video is very intimidating. It's intimidating to get in front of a camera. It's intimidating to speak. It's intimidating to edit that content after you've recorded or created the initial materials. But I think understanding that these AI tools can help anybody build up that confidence, and that's one of the key things that we try to build our key value propositions around these tools is building the confidence to start creating video in the first place and to be able to edit that content basically just like you would edit a text document. So the way we see it, once you have that confidence, you can go from start to finish in creating a content, and we call it One Take. So if we're talking about executive communication or marketing communication that is primarily talk to camera style video, which we see more popular than ever, I think, today. And I think that's just going to continue to evolve. You essentially can get started with something that's not a blank canvas. And we want to try and help jumpstart that creation process through various tools. One of the tools that we just launched, for example, is the AI script generator.

So essentially everybody can now have a certain starting point to jumpstart their creation. And again, this goes back to leveling the playing field. We know that professional speakers have people helping them out with their content. And I think this is a great example of how AI can be leveraged to enable anybody to get started and to unblock that writer's block that we all get sometimes like, where do I get started? And again, this is just for getting those creative juices flowing, and we expect and we see that even when AI is used to help you create your content, it's just a starting point. It's just like a template to get you started. And once you have that script ready, that is part of that confidence building to get started and record your video because now you don't have to just ad lib it or you don't have to, one of the things that we've been seeing, and I'm sure all of us have experienced this that had to record video messages is having that little notepad on the side next to our camera to read off of. So we actually integrated that into the product suite with a built in teleprompter that you can position close to your camera.

And again, this again goes back to that key theme of leveling the playing field and building confidence. Teleprompters are used in huge production studios every single day. TV studios, your favorite news anchor, they use a teleprompter. Now anybody can use it and have their videos look more professional and get started with ease. And we're adding smart AI based features to that as well, which is super exciting. And once you're finished with your recording, like you mentioned, Neil, before, you can magically just remove all those mishaps that we all make, ums and ahs, which I say a ton of so I'm thankful that we created this tool.

I use it every single day, and it's just such a huge time saver and again, I think this is just the tip of the iceberg.

- [Nikolai] I find there's a lot of standard types of videos out there which AI is perfect for because anytime there's a pattern, that's something AI can start to replicate. And a lot of the standard stuff, standard types of videos you see, and a lot of the recurring effects and formulas, it seems traditional nonlinear editors are overkill for a lot of those typical edits and like whatnot. And then as far as like the accessibility you were talking about. Video production for a company that isn't primarily video oriented is like a huge undertaking, and it's multimodal.

So there's like audio, there's the visuals, there's post production, there's writing involved. I've seen that a lot of video tools have come out that have really tried to take each part of production and make it easy to do.

- [Zohar] I think you hit on a few points here that I think it's worth emphasizing. And like you mentioned, today the video creation and editing workflow is really fragmented. You really need to jump between a lot of different tools and products, oftentimes pay for multiple different licenses just to get a simple video out, a simple workflow.

And that's one of the key pain points that we realized that a lot of creators and SMBs and businesses in general, small and large, are facing. And historically, you know, Vimeo is known for its top notch, best of class video hosting and video player, but we really offer an end to end video creation or video experience platform rather to enable you to really go, to enable businesses to go from end to end through the entire workflow in one place. And it's all cloud based. So again, you don't need to install any software. You don't need to download, upload. It's all done in the cloud, and there's huge value in that as well. AI is actually not something that's new to Vimeo. We've been in the AI space for many years now. We actually have a pretty decent size AI team that's solely focused on AI tools whether it's in video creation. So we actually acquired a company in 2019 called Magisto which automated video creation. A lot of video analysis, audio analysis, a lot of those different capabilities are already built into the platform in various ways.

And also on the backend. Like how do we make our platform more efficient? The creation side of things is just one part of it. There's a huge side, probably even bigger, of just the consumption. And I think we focused a lot of the conversation on creation and editing, but I think there's a lot to be said also on the consumption side of things. How do we leverage AI to extract knowledge and extract information out of videos to make the viewing experience more efficient? A great example for that is recorded events or webinars that are created and hosted. They're all recorded. You end up with a one to two hour long video. How many people are going to watch that piece of content after the fact?

So we're trying to think of ways, how do you make that piece of content more accessible to find the information that you want to, in two seconds, understand what was discussed in this video, what are the key parts? What are the most relevant points for me and how do I make that content searchable?

That's the way we're trying to look at it from. Video creation, video editing, making videos interactive, live events, virtual events, all the way through the repurposing, re editing, and consumption part.

- [Neil] That's actually fascinating because when you were talking about that Zohar, it reminded me when I wrote my first book, right?

I thought writing would be the hardest part. Probably turned out to be the easiest part. Then you get into the publication cycle, the editing, all that kind of stuff, and that was a ton more work. But then you have to do a lot of work to promote the book, get it out there, all that kind of stuff, and I think the consumption part, you're spot on.

But, I think most people, they think they have a feel for that. These tools you're creating like with the help of the script writing, is that all focused then towards I'm going to help you create the best possible content for your specific audience. So that as you go through the entire sort of steps here, you'll be best positioned from editing through consumption?

- [Zohar] The fact that you have everything in one platform, you can really leverage a lot of that data to make decisions later on in the process. So for example, if you're hosting a virtual event or webinar, you know, we can use different data points including AI to understand what were the key moments in that piece of content to help you later on in the post production editing process to create highlights, shorter clips, the aspect ratio to fit the right platforms. So there is in addition to the benefit of having everything in one place under one subscription, there's also a lot of benefit in having all these different tools know how to speak with each other and have a shared metadata that can be leveraged to different, in different stages in the process.

- [Neil] That's amazing. I did not read it off a teleprompter, but the fact that you could take whatever event or video you're making, and it's smart enough to understand that, hey, this 45 seconds here would make a fantastic video grab.

- [Zohar] And even when you're asking the AI to generate the script for you, you can ask also for the duration, if it's short, medium, long, the tone of voice that you want. If you want to be confident or funny or casual.

And that's, again, that's just the beginning. We have so much plans, some of them coming out very soon as fast follows to the stuff that we recently launched. We're already getting a ton of feedback about what users are looking for. How can we make this better? How can this become more efficient?

Again, I think we're just getting started.

- [Nikolai] Kind of feels like there's two classes going on right now. You have the first class of tools that are like about what we're talking about, like extracting like the best clip or doing auto transcription or doing like basic edits.

And then you have like this more emerging class of tools that's like full generative, like just generating stuff out of nothing. We're starting to see like AI avatars. Actually have access to a demo recently. They are getting very good, they're indistinguishable from real life.

If you think about especially filmmaking and Vimeo has a lot of filmmakers on the platform. It's always felt a little traditional and a little verbose to me until recently. And it's like only now we're starting to really see a lot of cutting edge tech finally start touching video production and film production.

There's the Hollywood strikes going on as well. So I guess I'll jump to that question of just what are your thoughts on those strikes, and then I guess the broader point of AI in film production and video production, the kind of impact that, especially the really cutting edge when we get to more and more generative with avatars and whatnot.

- [Zohar] When we're talking about filmmaking, I do think that this is another tool that filmmakers, whether it's writers, animators, artists, even actors for that matter, will need to embrace. Technology has been an enabler for creativity for years.

To create an animated film, an artist needed to draw every single frame to create those moving images. That's not how movies are made today thanks to technology.

I think with actors and copyrights, which by the way is also part of the strike, it's a very important question. You can clone somebody's physical features and their voice and train large language models to sound like somebody whether it's an actor or character, an artist, a music artist.

So there's, it could potentially be very detrimental if it's not handled in the right way.

- [Ryan] You talked about it leveling the playing field, which I think is right. It is directly in line with the value a lot of these businesses are going to get from learning these tools, being able to utilize them and be able to just like start doing video when before, like you said, it was very difficult or it's a bigger lift for organizations to adopt video.

That's how I've always known it, right?

- [Zohar] While it levels the playing field, it also raises the bar for what is quality. And what is authentic and what is creative? And I think that's really exciting in my opinion because it's going to require brands and creators to be original and think how you can continuously, improve and push the boundaries. And what makes content special. And if you think about 5, 10 years ago, video wasn't even you know that mainstream as it is today. And the definition of video was evolving. We saw vertical video and being introduced to adapt to the mobile consumption today. 15 years ago, or even less than that, it would be unthinkable to record a video in vertical format.

There were actually a lot of people who came out against that because it was consumed on computers, computer screens, for the most part, and you had those big black blocks on the side. We saw the evolution of text on screen videos, which were, they became super popular. Why? Because Facebook changed the way videos play in your Facebook feed and they auto played the video without audio. So it was important to add the text on the screen so people can follow the video while it's being consumed with no audio. And I think that kind of created that evolution.

We're still seeing that and how videos are captioned today. A lot of those TikTok videos that show your words as you're saying or emphasize those words as you're saying them, editing videos through the text. And I think it's just an evolution and we're just going to see how creators and creative people leverage all these tools that we're seeing introduced into the market to continue to raise the bar and continue the evolution of what a high quality creative video is.

- [Ryan] So that's a good point. So actually a question I wanted to ask you since we've been discussing this on the written content side of the world too is with the barrier to create written content dropping dramatically with the help of tools like ChatGPT and others, it's going to be easier than ever for companies to create written content for content marketing purposes and so forth. So on the video side, is there any fear that the ability to create videos more quickly or easily will dilute video, just the video experience? Or will it put more of an emphasis on quality of the content and substance of the content and things like that. Like how will people be able to stand out?

- [Zohar] Still the human creative element I think is something that it's going to continue to be important in the process. And if you have certain content that's being commoditized, it's going to feel that way.

And you're going to have to find ways to stand out in the crowd and I'm going back to the text on screen videos because eventually all the videos on Facebook started looking the same, and I think there were some you know after a couple years there was a bit of fatigue in the market for those types of videos and then we saw again another shift in the industry. So I think that's going to continue to happen. And we're seeing that with text also, in my opinion. Like a lot of the text content that we consume today, I think we need to take with a grain of salt because a lot of these, it was content farms being generated by maybe low wage workers in some places and that evolved into now being generated by machines.

But still for the most part, we as consumers need to improve our filters also and understand what is quality content and what is valuable for us. And in some cases, yeah, automated text content is great. In other cases, I think that you still need to, you need the human angle or perspective.

- [Neil] To a point you made earlier, right, around you can control, you get coaching on the tone, that kind of stuff. I think even from a content standpoint, we're shifting towards precision communication, right? You almost do precision communication in that, I may shoot a video but using these tools, I want to make it more fun and upbeat, maybe a little humor, because I'm targeting this video for marketers, but the exact same kind of information, I'm going to use these same tools to create a more serious kind of down to earth video, even though it's like basically the same basic type of information, but for lawyers.

So now I can suddenly create multiple versions of the same video tailored for these different audiences. That sounds like it's not just the quality thing. It's going to be like this precision communication.

- [Zohar] A hundred percent. I think that tailoring or customization of content for the viewer is going to be a huge part of what we're seeing when it comes to generative AI.

Think about translations also. Think about the way we educate, we consume, the way we educate or teach or train, whether it's students or even the workforce, how efficient, how much more efficient that could be.

- [Ryan] This has been a fantastic conversation, really interesting stuff. Just the fact that a lot of the stuff I've been learning through Nikolai, so to have somebody on that is active in this world building these super advanced tools that are making something that for a lot of people did not feel attainable, super attainable now, is really cool.

So Zohar, thank you so much for taking the time. Neil, I want to throw it over to you. Any last questions or any last thoughts on your end?

- [Neil] I think this is fantastic. And I think that for all the people that are creating content or aspiring, you should definitely check out the tool suite from Vimeo. And Zohar, if people want to learn more about what you guys are offering, what's the best place for them to learn?

- [Zohar] The best place is to go to vimeo.com and check out all the different tools that we provide. I think that the Vimeo of 10 years ago is very different than the company we are today. And whether you're a creator, an SMB, or even an employee at a large organization, we're really all about making the end to end video process easier and more authentic in a human way without breaking the bank or hiring a production team or agency to do it for you.

Anybody can do video today. And we're putting out high quality, immersive content creation tools really in the hands of anyone, whether it's a communication team, a people team, a marketing team or even the executives at large companies who we all communicate through video today.

I would leave you with that thought and of course, check out our One Take video creation tools that leverage AI to help you in that creative process

- [Ryan] All this stuff is going to be linked up in our write ups and distribution because we want to make sure our audience really gets a chance to check it out. Big fans of what you're doing and as this continues to evolve, we'd love to have you back or another member of the team back to just talk more about the evolution of AI and video because I think a lot of the businesses out there that are listening to this are going to get a lot of value out of bringing this in to compete when it comes to that content marketing, thought leadership, just brand awareness levels that they weren't able to compete on before because they didn't have the resources or the budget or the time, so this is making that possible now. So thank you so much for your time and look forward to getting this out to our audience.

- [Zohar] Definitely. It was my pleasure. And thanks again for having me.