Маркетинг на съдържанието и социални медии в ерата на изкуствения интелект

0 от 9 урока са завършени (0%)
Дължина: 15 минути|Трудност: Стандартен

Лекции

Автентичност срещу автоматизация: Състоянието на YouTube през 2026 г. от Лариса Иванова

Слайдове от лекции

Относно оратора

Запознайте се с оратора

Лариса Иванова

Лариса Иванова има над 8 години опит във видео оптимизацията и стратегията, подкрепен от степен по психология и солиден опит в маркетинга. Тя е основател на консултантска компания, посветена на подпомагането на малкия и среден бизнес да се развива чрез YouTube. С допълнителен опит в управлението и организацията на събития, Лариса предлага уникална перспектива, която съчетава поведенческата наука с дигиталната стратегия, което прави Video SEO нейната истинска суперсила.


Препис на лекция

Представяне на лектора

[00:00:00] 

Welcome, Larissa. She’s our next speaker. Her presentation is about authenticity versus automation, and she’ll be sharing a really fun experiment that she did. She has over eight years of experience in video optimization and strategy backed by a degree in psychology and strong background in marketing, which I’m sure you have all heard of, if you have ever been to a conference in Bulgaria.

And she’s most recently the founder of her own video SEO Agency, Avenara. She’ll tell you all about it, but it’s helping small and medium sized businesses grow through YouTube. And she has additional expertise in event management, which we love. Thank you so much for bringing it into the community and the unique perspective that blends behavioral science, digital strategy, making video, SEO, their truth.Superpower. I hope I did it justice. I’m sure I haven’t, but [00:01:00] it’s okay. Fun fact about Larissa before she starts is that she’s professionally trained to fight with a katana, which is a traditional Japanese samurai sword. So yes, take it away.

Lecture by Larisa Ivanova

Okay. Hi guys. This is my, basically first time.

Yes, thank you. First time on stage and speaking. Bear with me. Yeah she did introduce me enough. So I’m not going to go through the introduction about myself, but today I’m going to talk about authenticity versus automation on YouTube specifically for the next year. And I’m going to have a little bit of a different approach through human behavior and psychology.

So this is the About Me section, which I’m going to skip. And of course today we are going to go through four different stages. Oh, I’m going to get in trouble with this microphone. Okay. Is the AI content bad for YouTube as its own? [00:02:00] A year ago, two years ago, many titles became really stressing the people out that YouTube is going to delete every single bit of AI content out there.

So if you make the disgrace to upload AI content, YouTube is going to delete you and basically your family. So that didn’t happen because the idea from the YouTube site was they just didn’t want to have said to get the the platform entirely full with AI slope, which is thing we all have seen by now.

And also they want to protect the creative people on YouTube who basically this is their job and this is their work. So the idea from the YouTube side was, no, okay, guys, upload ai content. Try to use AI content, but just make it, make sure that your content is not trashy, your content [00:03:00] is not meaningless.

If you are have a valuable videos, valuable ideas, go and do that. But of course right now there are a lot of videos uploaded every single minute on YouTube, and this is day by day and it’s going to get a lot more, of course. And the attention span is getting lower again by day. So right now we have a lot of hours of videos uploaded every single minute on YouTube, which is a problem.

So the AI ban from YouTube and demonetization was a thing that YouTube wanted to implement just to protect the platform from meaningless stuff. Because I really don’t believe any single title I see. I really like to check everything I read online. I created an experiment, and you can see here that these two different videos, which looks exactly the same, are not because one of those videos is me in my living room and [00:04:00] another video.

The second video is my AI avatar. The idea of the experiment was to see what will happen if I upload to exactly the same videos on YouTube on the, exactly the same channels to see how YouTube will react on that videos. So what was the environment to exactly the same channels? The description of the channel was the same, absolutely everything was the same.

The title of the videos I’ve uploaded because I didn’t upload only one are absolutely the same. Everything was the same. The thing I was really adamant about was I didn’t want to promote those videos. I didn’t want to share them. I didn’t want to push the algorithm further. I just kept my mouth shut and let the algorithm do its thing.

Of course, one more thing here. The topics of the videos wasn’t something that would have the opportunity to go viral. It was something that will be similar to what would small business upload as a YouTube video, educational YouTube video. So [00:05:00] these were the results. The first column are the assets.

We are going to see the video RP one, video RP two. This are my real person type of videos. And the second thing is the AI videos, which I uploaded three from the two test objects. By the impressions, you are going to see that YouTube really pushed a little bit further. The AI videos, they got a little bit more impressions.

When we talk about the views, though, it’s. 50 50. Basically one of the videos, the real videos got a little bit more views and one of the AI short videos got a little bit more views. When we talk about the likes, here is where the things got its turn, because in the in the beginning, in the like section when we talk about the real videos, there were likes at all because people liked the videos.

And in the AI section we have two dislikes, [00:06:00] which is something that didn’t happen that much when I uploaded another videos anywhere else. I got really clear picture of what’s happening. Of course, the comment section is the same. The next thing here you have for the presentations just to see the real results the thing that was in the table.

So what’s, what was the conclusion of everything in the this experiment, this small experiment? U YouTube doesn’t have any problem with AI content people do. And yeah, this is my tools that I used. You can check them later, but why people do have problems with AI content. Because of something called uncanny valley effects, which is the psychological repulsion of things that are near but not quite human.

And of course we have all see that type of content everywhere. And most of the time our reaction of that type of content is “ugh, that’s AI” and it’s some kind of challenge [00:07:00] with ourself when we scroll through a TikTok or any other platform, when we see AI video to point it out. Of course it’s normal to have a feeling of something unsettling when you see on AI content.

And that’s because your brain works over time, behind the scenes. You don’t even realize when your brain gets the few things in the picture, not quite that are not quite right. Why does that happen? There are really cool and interesting experiment done in the California University of San Diego where one neuroscientist, I will butcher her name.

It’s on the second next slide, so I’m not going try. But we have a test, which is a few people, a lot of people are going to be shown three type of videos, which are just a second, which are, the first one is absolutely a robot, which is doing some things on the [00:08:00] video. The second thing is an Android, which is a human looking robot.

And of course the third video that the test objects were seeing was on a real person. And you can see by this picture exactly that. Basically when people are trying to figure out what the heck is happening, our brain flares up. And the interesting part here. Which is, yeah. The interesting part here is that you have different places, different sections of your brain.

And I’m not going to go into deep detail here, but you have on that picture here you have two specifically flared up places of your brain when you are trying to figure out what is happening. One of this, which, which here is our frontal lobe, which is the part of your brain that works for solving problems and checking what’s happening.

And of course, the parental lobe per parental, sorry, which the part of your brain that [00:09:00] works for, from a perception point of view, this is the part of your brain and exactly this part here, which is working specifically on understanding the world around us. So seeing this you can see that when we see something that is usually not quite right, we have our brain flares up.

And this is really important and it’s something that evolutionary came from millions of years ago because whenever our ancestors saw something that looks like something else that most likely were I don’t know maybe you are his dinner or it’s dinner. So it’s really not careful to trust your eyes most of the time.

So this is something that evolutionary has set with our brains to protect us. And right now it helps us figure out the, the culprits in our daily work and it’s something absolutely normal. So the solution is to try and train the AI models that [00:10:00] they basically have said, Polish the content we are providing on YouTube, on the, or the different platforms to be almost exactly like something real, because this here, oh, again, one of the parts is AI version of me.

One of one is not it. I’m really interested right now to have a bet with you which one is which, but let’s talk about this later. Okay. Of course the tools you can use, we don’t have a lot of time. I’m not going to go too deep in that, but I have mentioned a few tools that most are most used right now when we are talking about something with AI content.

The first one is Hagen because this is the tool I use specifically for my experiment. It’s really easy to use and really easy to manipulate. The second thing is the script, which is not for creating AI content itself, but editing video online. And the third thing is at least, which is if you have used Mid Journey and open ai, this is something like their mutual [00:11:00] child.

Because using Art List, you can use the model with what’s the videos are going to be made. And yeah. And here I have a little bit of hot tip using card list because you have different models and using Art List is the magic of using Art List is in the prompting you use for it. And because the prompting is really interesting part and a lot of work to do to learn how to create technical part of prompting.

I have given you here a cookbook that you can teach. Ai, open ai how to create your own prompt. You are going to receive this link. The only thing, put this link in, learn to tell him to learn how to do your prompts. And afterwards, you can translate your prompts like with Google Translate. Okay?

This is the examples. I’m going to go through them. And the final advice I have for you, whenever you are crying, you are not [00:12:00] crying, trying your AI Avatars in your teaching the AI model to create your AI Avatars use as little as possible. – e.g. Jewelry on your face. Glasses everything. Pearls are the biggest enemy of the AI models.

When you try again whenever you want to teach and provide a specific AI avatar, try again and try again because you are not sure which version would be best. And of course, trying is better. Stretch limits every time you can. This specific advice here is for every single new tool that’s, that comes up because tomorrow will have another one, and the next day and another one.

Basically put your hands through it. Take it to parts. Learn how to use it. Don’t be afraid of it because if it sticks and it’s stays on the market for a longer time, you benefit from knowing how to use it. And of course this is an advice from the [00:13:00] marketing from the business point of view.

Whenever you create online content for YouTube be the best answer out there, whatever you’re using, whatever you’re using, everything else just creates meaningful content. Yeah, I don’t know, do I have time? I have? yay!. Perfect. Oh, perfect. Yeah. And of course, because this is my passion to, and Lazarina’s passion, we have created, we are working right now on so cool.

Course, course not curse. We are not cursing anyone on scale scalable YouTube, a SEO course. So me, I’m going to in this course, if you’re interested, I’m teaching you how to work with YouTube and how to scale and automate the processes. This will be available soon and you, I’m gonna receive more information soon about it.

I think in the beginning of December, this will be available yep. That was me. I do have my [00:14:00] personal video, a SEO consultancy I work on in the moment. And my focuses are YouTube strategies, video strategies, and basically working with people, trying to help them navigate YouTube as a tool because it is a precious tool.

Yeah, you can find me, you can contact me and if you have any questions, feel free to ask.

bg_BGБългарски