AI Masterclass at Masters’ Union | Must WATCH for Everyone Curious About AI For Content

I've actually never spoken about my thoughts on AI for Content publicly in one shot ever so my first golden rule for everyone here is Media does not make money here's my point if you're a bad writer your value is $20 a month all the media you consume at some point is going to be synthetic completely synthetic all right so what I productized myself now I can produce 100 videos a month thousand videos a month a million videos a month about everybody topic ladies and gentlemen thank you for being here thank you for uh you know firstly Master's Union for having me here um I've actually never done this slide deck I've actually never spoken about my thoughts on AI for Content publicly in one shot ever right this is probably the first time I'm doing it um and I think I spoke to the audience before we start rolling the cameras and the average is about 18 to 21 22 so I have to keep this simpler the deck might be slightly more complicated I'm going to try and pull it down uh for those in the audience that are listening this is going to be a lay maniz version of you know my journey with AI and content hopefully it's useful hopefully you take something away from it um I want everybody to understand one thing right uh for me AI is only as useful as the outputs right as what I can use it for I don't actually care about um which tool I'm using as long as the task is done can we all agree on that right actually this audience here this specific audience you don't care model model you're like this tool can do this right fair you want to use it for a for a task you want to use it for the end goal of whatever your your dream is right either to make content or to make a video or you know maybe you want to write an email right you're just like I want the best tool for this job and I want it to just understand you know whatever I'm telling it to understand so I want to tell you my learnings around AI for Content uh because we use a lot of that I'll also tell you what I'm building uh I have a very very firm thought around um you know a few things in my life one of them is the only way you can truly learn something is by doing it so over the last one year we have created so much content around ai ai tools all the latest tools and me and T when we started doing this we had one rule which is we will play with every tool if it's a tool we'll use it if it's an open source repository we'll use it if we can make our own version of something we can fine tune it to build something that is useful for us uh that we can display live that people will enjoy we will do it if you need to clone Shah ruk Khan's audio to prove a point we'll do it right so um when you start with that thesis ke I will try everything however hard it is to try it at that point of time you learn a lot and you also learn about opportunities that other people can't really see and we have used some of those opportunities right and we have seen that some of those opportunities are far larger than people think they are so hopefully you get to learn something from this so this is my thoughts on AI for Content first about me okay I studied computer science um I was a I studied in manipal Institute of Technology I studied computer science when I was 18 or 19 years old I studied for 4 years so I'm an engineer I raised Venture Capital very early uh in the year of college I decided to start a t-shirt company so uh I I was at your Drop Shipping challenge yesterday and you know some of the people they were selling T-shirts and for me it's like Nostalgia because you know so many years ago 13 12 13 years ago I I was selling T-shirts because I was like trying what is there to lose I was an engineer and at that time the cool thing for engineers to do was to sell t-shirts I don't know if you guys know this I said let me try so so but that's not what I raised capital for I raised capital for a recruitment startup called job SP because you know when you're studying engineering and when you really are passionate about building things especially when you're passionate about writing code um you build software and we saw zomato and we said we should build something like that for recruitment so we build a company called jober um that's why I know so much about the recruitment market right A lot of people are like what what do you know about job market it's like uh I spent a few years building a business there so I have some insights uh specifically we worked around fresher hiring so I have a lot of insights around fresher hiring um then over the last you know 10 years beyond that story I've built and sold multiple technology businesses my most recent one uh was scenes we sold that to uh on academies graphi um I've been an entrepreneur for for all my life uh I started actually with game design there was a very very old game engine called mugan you probably wouldn't have heard of it I had a computer when I was early I love playing games games I said I want to make games I start you know experimenting with 3D so I've been a nerd frankly uh but over the last 10 years I've learned how to be a nerd but also make it simple enough for the average person the average 18 to 20 year olds who for are not nerds to understand to use many of the things that I love using um yeah but today we run a AI services company uh it's called AOS uh Labs actually and uh AOS my parent company naos Labs is one of those companies that sort of you know we we have clients like Amazon Prime video denck basar Times of India Sports Kea and more we also built lots of AI tools we built Gard box Alpha CTR autocode Pro we have more than a million users on our AI tools we are profitable um and we learned a lot building those I have some insights about that as well hopefully I get to share some of it today and what most of you know me for is I create lots and lots and lots of AI content right we try to be the most legit authentic source for AI content but we try to do it in a way that is simple enough to understand so we have to dumb some things down but it's our job and we try our best to do it uh even though some of it is a little bit complex um but this is about what I've done so that is why I do it the main reason I do it is because I think that we are going to enter might be a pipe dream but I think we're going to enter an era called a post artificial general intelligence era where AI does a lot of what we call econom I Ally viable work but we will still be very valuable okay and I will come to why we will be valuable over over the next few slides I want to document every moment of it right for me it's like the world is changing um into something very different right I have you all seen the videos about Vision Pro a lot of people wearing the Vision Pro running around walking around you know it's strange right like today we walk around with a phone and it's just considered normal everyone's like it's considered normal but you know there was a time where it wasn't it was rare to see somebody with a phone especially when I was young and today just become normalized I guarantee you things like The Vision Pro is going to become normalized and you forget how did we go from pre phone era to phone era it's almost like none of you imagine none of you remember a world without Instagram and Snapchat but I do and I feel like it's important to document the world we're in right now because 10 years later you'll forget this world so I want to document all of it coming and I want to build lots of entertainment games toys um and I want to build and sell really cool stuff right that's my philosophy I want to build and sell really cool stuff and I don't want to raise money again because I think I'm far too experimental for it and I think the world is changing so fast you might do something today and a year later it might be worthless so it is we are starting to enter a very risky world and one of my friends AE AA BD um who runs super team in India he's like now he's like apps are like content right you'll be popular for a few years and then you die especially in consumer which I really enjoy so I've learned that I don't want to raise money again I want to do everything I do bootstrapped um awesome so I want to tell you why content our first learning was this wave of generative AI really disrupts content does anyone agree with me here okay give me an example yeah so like uh if we like we can see lots of videos like especially sat videos about Donald Trump Joe Biden and like Barack Obama having conversations which are the content is AI generated like everything is AI generated yeah and it's quite fun as well yeah I mean you know when when AI first came out right few years ago a lot of the people especially nerds right like me on Reddit said okay this is going to disrupt recommendation engines this is going to disrupt uh you know uh what products are shown to you and sure it did but but all that happened behind the scenes hidden what really disrupted was you're right Modi singing some song uh you know and and people cloning people's voices and faces and all that chaos right so we learned that AI really diss content and there's probably the potential to build a Marvel like movie in India today with just an iPhone or a regular phone so I want to document how you can actually do that over this video uh I mean over this presentation it's a content creat dream you enjoy creating stuff but it's really expensive to create a movie right it's very very expensive ask the people who made ADI purush so it's a dream that we have the ability to do it as common citizens as common individuals without having all those resources uh and most businesses today come to us even when they come to us for AI needs right like I told you we run an AI Services Agency they still come to us for Content needs it's like aamp can you guys do this campaign with AI so that's the kind of thought process uh that people come to us with so we really truly understand that the need the use case for businesses even though we can solve Finance use cases and you know chat with PDF use cases we feel like this is our Forte and this is something that we are good at um we recognize that content brings us leads and users you know I I keep seeing this comment sometimes on my uh YouTube right why are you doing content if you're running a business anyone has a good thesis as to why you should do content if you're running a business distribu distribution okay what does distribution mean okay launch ation they will be my first user and pay me $10 I will having $100 sure I I think people trust people pay money where they trust people if you trust somebody you pay money right um and I think content really allows you to build that channel of trust because you're sort of talking to them hundreds of times without expectation and then at some point they're like I want to do something with this person right and for me it's kind of been the same right everyone that I watched on YouTube that I really loved I now have the opportunity to work with some of them so I understand what it's like so it brings us leads it brings us users um it's really the power law for us uh we start creating content in-house we start for fun by the way we said time pass it's like it's almost like people start a rock band when they're bored like I've seen some 30 35 yearolds they have a mid-life crisis and they decide I'm going to start like you know a band with my boys we sort of started YouTube the same way we we didn't really think it's useful or anything right um little bit of personal branding fine I I get it but we didn't really think it would be Val valuable now we do 100 million views a month across channels and platforms we have seven different channels run by seven different teams um so we we crossed 100 million views which is uh phenomenal considering they're all English views we don't do any Hindi or Regional content yet we will not do Hindi or Regional content right now because we feel like in a year or two YouTube will have automatic translation it will separate all the tracks the backgrounds music the foreground Etc and then it will automatically translate that into what a language of choice so all the creators today who complain saying y I'm only getting 20,000 views because I'm creating English content that's going to change because somebody who speaks Canada is going to use YouTube primarily in Canada is now going to see all your content in Canada YouTube gets to expand inventory like this right so uh we do 100 million views a month on English content which I think is pretty cool and we have just at least for our own content there's a ridiculous amount of cost and efficiency savings with they I um but I do want to tell you that content does not make that much money um YouTube pays really poorly it pays you peanuts um so you have to build businesses on top so we see everything we do on the content side as distribution and that's very valuable but then you still have to build businesses and those are almost like two separate silos two separate businesses uh you know for for at least the way we see it because it's like the minute you start doing business inside your YouTube thing the YouTube thing gets disrupted people start focusing more on the business side Less on the YouTube side it dies so said it's got to be separate things that's your traffic Source you have a business um I'm going to put in some golden rules throughout this so my first golden rule for everyone here is Media does not make money okay every good founder knows this so the good Founders don't do media they see are why should I do YouTube it's not going to make money it's going to lose money I want to spend whatever x lakhs a month and then get no Roi back um but it's one of those things right it's like brushing your teeth if you brush your teeth today you are going to get nothing from it your teeth are slightly cleaner today you look slightly cooler today but you get nothing from it in the short term but if you do it over 10 years or it's like going to the gym right like you go to the gym today nothing happens like this a waste of time I wasted 1 hour of my time I felt a little bit good but I wasted 1 hour of my time but then uh a decade later you're ripped and being ripped gives giv you a lot of secondary benefits right which I'm sure some of you who have who are wellb built or go to the gym understand these secondary benefits that I'm talking about all right so when people ask me what I do and it's it's always very interesting when people ask me what I do right I've been an entrepreneur for 12 years I've built businesses for 12 years I only created content as a hobby for like maybe six or seven years right I was a Creator parttime um but today I'm a content creator like of all the things I do the thing that I really point and I'm like this is the power law you guys know what a power law is you know the power law is a power law is it's the 8020 rule right you know the 8020 rule 20 okay can someone tell me the 8020 rule so we focus on the 20% important part of the potion and out of the 80% VI we get yeah I think on YouTube or or content it's even more skewed it's like one of your videos will do so much for you that you're just like wow that's why I said media doesn't make money but I think at least in business a lot of business is luck that's my personal belief it's luck like you don't know which company is going to reach out to you and say can you do this for us it's pure luck and anyone saying I can predict how much deal flow I'll get next quarter unless you're like massive scale and you're a consumer business I think they're lying it's a lot of luck but distribution increases your surface area of luck today I wake up and I have no idea who's going to text me on Whatsapp I don't know how they get my number but some of the smartest and coolest people who I'm like like 5 years ago I was like wow it would be amazing to talk to these people they text me so I think it's it it it increases your surface area of luck and you can increase your surface area of luck to the extent where every day somebody's checking in and wants to do something with you and we are seeing that compounding play out in real time right um so that's why I'm a content creator because distribution is the power law content is the path to distribution you have to create content to get distribution in the first place AI makes content production significantly cheaper have you seen a shot of me that's completely automated some of you have seen it completely automated you know the view count doesn't differ so much and we'll talk about that in this after a bit right the view count doesn't differ so much but for me it is an ability to scale myself sometimes if I make an audio mistake great AI will solve it so if I was to spend and I don't spend too much time doing content i' probably spend like 3 4 hours a week shooting the rest of the week is just me reading and thinking and like you know talking to people and getting insights right because at the end of the day when you watch content you want my insights you want my ideas and thoughts and you don't want to be ideas and thoughts you want me to put in the work talking to the people learning about enough about the world so that you I'm able to distill it for you right you want me to play with all the AI tools and you want me to try 50 different things and find out 49 cases where it doesn't work and you want me then to distill those thoughts and insights for you so I spent the rest of my week screwing around so that that 4 hours that I put shooting content is very valuable for you right so AI makes content production significantly cheaper it allows me to scale myself in a way and what I've also learned and the reason I'm a content creator is you can partner with amazing teammates right which distribution will bring you because the minute I put up a story you get really smart people applying you get users which distribution brings you and you gain trust which distribution again brings you so we have some really smart Engineers for example who work with us why do they work with us when they can work at any other company cuz they're like I know if I work here half a million users will use what I build a million users will build use what I build it may fail it may not work the product at the end of the day the business might go bust but they're like I know if I build this thing and all smart Engineers have realize this right you build something nobody comes if you don't have distribution but you partner with somebody like you know it's going to get some number of you know views and from there it's pmf right either the product works and the market really wants this thing and therefore you're making money or it doesn't work and it dies but at least you got a shot of day because there's a graveyard of products built by people nobody cares about them and after many years smart people have realized the main thing is the thing I build I want people to at least see it and use it right so for us content is the power l that's why of all the things I can be the most important thing for me to be at least for the next couple of years is a content creator that doesn't mean I don't know how to run a business we've been doing it for 12 years we're a fairly large team now but this is where most of my focus and attention is now starting to shift very different from 6 months ago where I was completely focused on the business um so that's Golden Rule Number Two right and the reason we spend so much time talking about content and creating content the average 18-year-old in the United States today will spend 93% of the remaining time of their life free time outside of school in their life scrolling you to spend 93 the average Indian spends like 4 hours a day on video Tik Tok long form so why in India isn't there a video focused company yet I mean everyone who's doing it is doing like an agency model like right so we think there are more than enough people who are spending time on their phones look at your own behavior the most time and please anyone who's not doing this you can put your hand up and sayon you're talking nonsense you're spending the most amount of time per day on your bloody phone and you're watching video you're not reading text it's boring to read text it's cognitive effort but a video is fun somebody else is put in the creativity somebody else is put in the time either on Netflix or YouTube or whatever so the amount of time people are spending on video is staggering and they're bored they're spoiled for choice there's enough and more content you can watch you are bored you are spoiled for choice so we have to differentiate we have to do something better and there's a nice Sam Alman article I just copy pasted screenshot of the entire thing it's very important uh especially when you're doing something that looks like a joke in the beginning right content creation at least three four years ago look like a joke by the way most content creators who are doing creating content people like YouTube but I want to read this out for you I'm just going to read it word him a Founder recently asked me how to stop caring what other people think I I didn't have an answer and after reflecting on it more I think it's the wrong question almost everyone cares what someone thinks though caring what everyone thinks is definitely a mistake and it's probably important caring too much makes you a sheep but you need to be at least a little in tune with others to do something useful for them it seems like there are two degrees of freedom you can choose the people whose opinions you care about and on what subjects and you can choose the time scale you care about them on most people figure out the former but the latter doesn't seem to get much attention the most impressive people I know care a lot about what people think even people whose opinions they really shouldn't value a surprising number of them do something like keeping a folder of screenshots of tweets from haters by the way I do something similar like I I have track of generally of all my hater now now maybe not not so much but um it's like what makes them unusual these people is that they care about other people's opinions not today but on a long-term Horizon in my head I'm like we'll see in 5 years we'll see in 10 years because I've been around for 12 right and I've seen the world evolve and I've been right about a lot of things I've also been wrong a lot about a lot of things so that feedback loop I've been through one feedback loop and I know all the things mobile phones look like toys in India I know how parents treated mobile phones I know how schools treated mobile phones right I remember and now suddenly it's the most valuable thing you can do you do 90% of your work on it wow right so as long as the history books get it right right they take some pride in letting the newspapers and other people get it wrong so you should trade being short-term low status for being long-term High status which most people seem unwilling to do a common way this happens is by eventually being right about an important but deeply non-consensus bet which is your contrarian about something you believe in something that other people don't and you're right about it 10 years later that's the only way to do well right so for me content is this right where even now it's a lot of people think it's not as valuable building the business is more valuable I disagree with them and it's it's like 6 months ago I would have disagreed with me today so we are learning a lot about this right now I want to get into AI okay what we're here for uh I've told you why I'm doing it and I've told you where we use AI primarily it's about content um and not only do we create content about AI we also use AI to create content it's like a two-way street um I just want to talk a little bit about neural networks right the inspiration for neural networks actually came from biological the biological neuron um but actually that's not what ended up happening right like I don't think today's neural networks really mimic biological neurons to the extent that people originally thought it would a couple of early successes in neural networks exaggerated um the potential of neural networks a lot of people thought this is going to be really crazy in the' 60s and '70s uh but it turned out it took a lot lot lot longer and I'm going to keep it simple for you I'm going to show you a quick neural network um so as you can see that that's the input and I don't think you've ever seen this you I don't think you ever know what's happening on inside the Black Box I think we're getting more and more complex neural networks as time goes on uh but I want to show you an image neural network because I actually spend most of my time with diffusion models rather than llms so we really focused on images and video U so as you can see at the initial parts of the neural network it's learning small things like diagonal lines contrast as we get to the inner layers it's learning more intermediate Parts what's an eyeball what's a nose what's a eyebrow right and then as you get to the outer layers it's learning full faces this is very important to understand right because we actually have very low control of what is learning at each stage we only know how to add more layers right but we have very low control of what it learns at each stage here's another example right so as you can see learning facial features that's the data set from from the input layer to the hidden layers to the output layer you can see that it learns by learning Primitives and then over time it's learning more complex objects right I this is a younger audience so I want to keep it simple just remember that the thing is learning it's learning parts of the data set we have little control over exactly how and why and where it's learning and what it's learning but we do know for this audience there is something going on in that black box where it is able to understand it is not just a bunch of rules it's not just a plus b equal to C right where it's transmuting your internal data in it's transmuting your prompt in some way that you know is magical there is learning going on inside right so I want to stop here and there's an interesting question that Yan Lun uh who's the chief AI scientist at meta had put out once on Twitter and I want to ask you a question is a cat smarter than a dog yes yes how many people think cats are smarter hands up okay how many of you think dogs are smarter some of you don't know have no response cats and dogs two of the most popular pets worldwide have significant differences in their brain structures cats have 250 million cortical neurons dogs have 530 million cortical neurons so dogs have roughly the 2x the neurons number of neurons as cats okay so dogs have the capability for more complex mental processing compar to cats right a bigger neural network a big bigger set of neurons has more processing capability than a smaller set of neurons great uh you know I I can get behind that but all of us are smart enough to know that there are certain things that cats do better than dogs and there are certain things that dogs do better than cats the point I'm trying to make is that cats can probably hunt much better than dogs right because with a few neurons which are more tailored towards that specific activity they are able to do that you know hunting pieace much better than dogs right now there's some other dogs that can probably hunt better than cats right specific types of cats so can we agree that what the neurons are doing is more important than the number of neurons true but there is one very specific thing that dogss can do better than cats right and you will see this again and again and again which is the concept of theory of mind it's the ability to attribute mental States beliefs intents desires emotions knowledge to oneself and others and to understand that others have beliefs desires and intentions I'll give you an example uh this is a shake right let's say I'm sitting here and I keep the shake and I go outside and I come back an hour later and the shake is gone how am I going to feel huh bad okay what am I going to think someone took it okay any other thoughts Dr it who drank it maybe I lost it maybe I kept it somewhere else what are you all doing you're all stepping into my shoes at this point right you're thinking like me and theory of mind in a way is the ability to think about other people's thoughts desires and beliefs and actually dogs have displayed a much higher ability to have theory of Mind than cats right so one of the cool things for us for a very long time and one of the things that really blew me away with Chad GPT right or gbd3 the first GPT that I uh sort of started commercially talking about um you know on YouTube and that is that it shows some theory of Mind chat GPD can actually be like okay if you left your protein shake there and you went out and came back here's how you would feel you would feel anger and frustration and all those things right so it was able to step into the shoes of how I think and it was able to extrapolate this from just training data of just text so that's very exciting for me but the point I want to make here is all neural networks are geared towards different tasks so the first misconception that most lay people have is that and I saw this at some point right chat gbt is the one responsible for neural networks somebody said sorry chat gbt is the one responsible for deep fix somebody said this on the internet I think uh you know the government said this right that we'll talk to chat GPT and we'll tell them don't do deep fix so I want to make it first and foremost very clear that different neural networks are Geared for different tasks the hope is that one day we have a neural network that does everything but right now we're not there and it's almost like you have to choose use the tool for the job all right and it's so weird because you can edit a video you can paste a video you can like cut a video into multiple parts and ask chat GPT how should I paste the video and it'll actually do a good job so it's helping you in video editing but not in the way that you want so you need to know not only the tool for the task the neural network you need to use the model you need to use for the task but you also need to know how am I looking at the task right do I want the full video edited that's a different thing do I want just uh you know the pacing of the video somebody give me the pacing of the video that's a different task right so you need to know the tasks and I want to quickly run through some math right to tell you about one very important rule here there are 200 million chat gbd users chat GB is a neural network for or a set of uh models for text let's say text generation roughly have you guys heard of mid Journey yeah what does mid Journey do images 20 million mid Journey users what's the world population 8 8 billion number of mobile phone users what's the Gap I just want to you know make this very very clear okay there are 8 billion people in the world 7 billion people have mobile phones so the penetration of mobile phones is pretty high can we agree on this it's pretty high what what the hell is the penetration of chat GPD nothing very low what is the penetration of mid Journey it's like it's negligible so when people ask me why I create content and sometimes create simple content right and dump things down like bro first let's get to 7 billion users right let's get here let's let's improve the penetration because we talk a lot about a lot more than chgb I'm like imagine mid journey and chgb are two of the most popular when we talk about more niche tools smaller ones I'm like do you know how few people have seen this how few people know about this so we want to increase understanding awareness and penetration of this stuff because we need to get to Mobile phones really quick because when this comes and it's already coming the way it will change the world like the way mobile phones have changed the world is going to be rapid and you want to be on the site that adopted mobile phones first when everybody told you Q scroll mobile phone mobile phone any of your parents said that you're not sleeping well is because of the mobile phone neighbor's cat died it's because of mobile phone stomach ache mobile phone headache mobile phone right we demonized mobile phones for so long I remember I I know this right um when I was sitting for board exams I didn't have a I mean I had a computer and my parents just shouted me for using the computer I'm like I'm just using a computer I'm I'm reading stuff I'm learning stuff they're like no study books so we keep demonizing technology and we and even now schools are like not allowing you to use chat gbt and I've seen a lot of this nonsense happen um I'm like you're doing the same thing again you did this with computers then you did this with mobile phones then eventually you allow it no today you are allowed to carry your phones here right yes you're allowed to carry your phones in most colleges these days cuz people like like oh it's Safety and Security and whatnot right but the point is you'll demonize it for 10 years you'll say it's a bad thing and then you'll realize that's where most of the money is going to be made so this time I'm here to tell you looking at the last 12 years don't make the same mistake I will increase my goal is to increase so much penetration of AI awareness in India that people don't make the same mistake they made 12 years ago with mobile phones that's why I get angry when schools and colleges say don't use chat gbt I get that you first have to understand the subject matter matter and then also use chat GB I get that I understand the Nuance but please don't say no none of this ever you can say this first then this but never say none of this that's my job and not just JG right there's so many other tools outside of mid Journey that nobody has heard of so the Golden Rule here is we are early you are so early if you use chat GPT you're so early if you use Bard you're extremely early if you use mid Journey dude yesterday I came to your Drop Shipping challenge challenge right all of you had made covers and packaging some of you had made posters and I asked everyone why is this why does this poster look like this when it could be made with mid Journey you don't have to you know steal somebody's art technically right like you don't have to copy paste somebody's art if you're anywhere doing that use mid Journey right why what is the okay what held you back from using okay for about mid Journey there's so many image generators that are free what what prevented most of you making posters from using one of those tools yeah doesn't give the type of output that doesn't give the type of output that we want not the right yeah I don't even think it's about prompts right I think it's more about efficiency part the efficiency in the sense most of the time if you put the correctly sping are correct but when the time of outcome theing AR correct the logo is incorrect correct so so you're trying to make mid Journey do text right or joury I talking D and you're trying to make it use text so that's the thing right and that's what we went back to different neural networks do different things and your job is kind of now synthesis putting all this together because when you take somebody else's design and just put it there that's an easier way to get a lawsuit right with mid Journey we're still unclear but the point is um it's like everyone knows these things exist but you're not using it on a daily basis whereas I know people we are using it on a daily basis and it's helping us so much and it sometimes it confuses me I'm like do you not see this the point is if if people are using AI on a daily basis and it's actually valuable to us and our clients why are you not using it you're the young kids you're supposed to know this stuff better than most 4050 year olds and 4050 year olds come to us and say can you help this this department use this so it's like staggering to me right so I want to go one by one and hopefully give you a refresher on what you can use today that is useful in your workflow we'll go from simple to Advanced let's start with writing okay for every business writing is important for all content writing is important do you agree do you not agree yes we think writing is the power law of content content is the power law of businesses these days but writing is the power law of content if you can write well you do an excellent job your content will work but writing is more about subject matter expertise than the act of writing itself do you understand what I mean can someone explain like how knowledgeable you are about the top correct so if I make you write about pubg have you played pubg okay if I make you write about pubg you'll always write a surface level article if you make me write about carpentry which I have no idea about I'll always write a surface level article like I won't understand the depth of the nuan in it noan by the way stands for more than meets the eye right what is the specific details of this that you know is not usually spoken about so so when it comes to writing we think that subject matter expertise is very very important so your ability to do research then add your own insights okay which is your thoughts that is not maybe people don't agree with you but those are your insights you want to write in your style because all of you have a unique style of writing can we agree yes Char writes like a like a BP employee it's like a US polite BPO employee so while AI can help you write some stuff it is a rising tide I will explain what I mean okay today if you are a bad how many of you think you are bad writers what does that mean okay what do you mean by you're a bad writer I have [Music] got it um got it here's here's my point if you're a bad writer your value is $20 a month chat gbt is $20 a month if you're a bad writer chat gbt is better than you so what it's what has happened is chat GB has gotten good enough to write as good as like the bottom tier of writers okay and over time it's getting better maybe not in a while but over time it'll get better but there is a ceiling to writing and the top most heing to writing is like some you know founder writing a report back to his investor or some business analyst doing a deep dive into a company where it's filled with insights and sorry but CH can't do that point I'm trying to make is CH is getting better over time but the cealing to writing like somebody like a Paul Graham the stuff he writes I'm going to read it it's valuable to me it's got his life experiences and insights I know it's not generic stuff that chbd has created so the point I'm saying is the skill ceing for writing is so high you need to be above a certain barrier for your content to do well right and it's about mining insights from everything that you've read learning adding your own life experiences showing an example from your own life and then getting to that ceiling okay so what we do is we hire great subject matter experts people who understand the space very well and we say for 5 days a week you just screw around you learn stuff about that whatever topic you're writing about you deep dive you go talk to the subject matter experts in this if you're not a subject matter expert right but make sure that whatever you're writing is your unique insights and your unique thoughts at the end of the day and sure you can use chat GB to structure you can use chat GB to cut it down into pieces you can use chat GB to pose questions that you want answers to and you can use chat GB to maybe you've done an audio with somebody you've done a call with somebody to summarize so we use chat GPD like a tool at the highest level at the lowest level when we do shots sorry but it's very easy to use sh to use chat GPT to just write your shot and we use it all the time especially when you do business or news or something like that and it's like a 30 second reel where like just Char the so that's what I mean by Rising tide simpler easy content where you don't need to put in too much work it's just regurgitating facts there chgb has gotten really good but the minute you want to get to the advanced level you need real human writers what we use for research is a tool called perplexity AI how many of you have heard of perplexity AI okay how did you hear of it so like I basically follow the founder on Twitter so like that's where I heard about got it so the founder has distribution yeah that's why you're a user got it most people use AI to create the end script we use it very differently we use it as a first draft for us perplexity is like a tree right where like go find out everything you can about this and then we'll do the legitim test legitimacy test later so perplexity cannot currently add its own insights you need a subject matter expert to give their opinions on something right you need somebody who does the thing on a daily basis to give their opinion on something finetune models right and we're going to see more and more models which allow you to fine tune based on your writing style and it's going to be one click online it's useful to copy writing styles and we have one of this that writes like me internally but we rewrite anyway one thing we've learned is and we don't use it as much in production is because after we write it we rewrite the script anyway cuz like can we compress this down even more is there a India specific inside we can add so we really think through the scripts a lot per video we save maybe 20 to 50% of our time especially shots it's like 70% of our time is really saved next audio background audio is important can we all agree one of the notorious things that on YouTube is one thing that YouTube is notorious for is copy strikes use somebody else's audio gone that video is gone so 25% of our videos need a revoice over maybe the audio is bad or we need to change some fact or we learned something 2 days after the video was shot where we like okay this needs to be changed now so we we we do a lot of voice re voiceovers or some brand will be like do it again um we do this all through AI background music licensing is a pain um and some of us videos like especially what my wife runs a um they do songs so we use tools like Sono a to make songs have you heard of Soo yes yeah so we use it to make songs all right so we use epidemic sound plus Uno plus RVC plus r stack plus 11 Labs so we've got our own stack right some for some audio stuff we save about 10% of our time per video and we don't have to re-shoot stuff for short audio changes and we will never reshoot a shot we have never re reshot a short video because we just like AI the entire thing the cost of making a catchy song today at least for a point where the audience is happy with it is zero and we can truly clone anybody's voice you want to hear it okay how many of you are fans of shuk okay have a czy if you send me a voice on WhatsApp you can actually put that as an input so lame let's see if it works that is so also we have a crazy does that sound like sh huh it sounds like shuk most people can't tell the average person in India can't tell a song of Who Lost His comedy but in Hindi and this is the [Music] out [Music] how do you feel what do you guys think one sec let me just zoom this in the point is you want to make a full song Today sure it will not be the level of I don't know an or something right but it's decent you guys enjoyed it you felt something great we're happy and you know how much it cost us how how much time we spent on it nothing nothing right so we are the the cost savings especially if you want to do music and we do music on some videos because it's it just creates a more entertaining package right there's a reason we love Bollywood it's pretty low next and like I said your job is synthesis to put things together which is avatars are nearly solved okay especially for shorts stock footage for really short scenes is easy to generate like we just spoke about animations for logo transitions are solved if you're doing an animation for logo or transition um there's a tool called animate diff that is excellent uh deep fakes are not so useful for us we have not I mean we do it as gimmicks sometimes but I don't know what you would use deep fix for on a daily basis because there was a point where we thought okay we'll get another actor and we'll put my face on him and that's how I scale myself but avatars are just easier and better here's an example okay so completely automated Avatar using H like people are more likely to interact with it sort of like w but in real life and the part is it was the Disney animators who played a big role in making these robots not the usual Engineers some work they used few very interesting techniques like stuff they used something called so I have a question have the views been impacted completely automated video anyone who doesn't believe it's a completely automated video okay it's completely automated sorry it's not impacct the views all right so what I productized myself now I can produce 100 videos a month th000 videos a month a million videos a month about everybody topic why should I be limited to AI I can make a News Channel with my face on it I'm asking you serious questions I I feel like people don't are not serious about this enough right they look at it and they're like it's a toy it's not a toy it works it pulls views this is one of the most shocking things for me this is completely automated it's not dipping My Views so why isn't everybody bloody using this someone answer me yeah so I have a question how much time did it take you when you made this and then it takes time no you'll have to do research you put it together you have to shoot 3 minutes 4 minutes to make it to make the verion No No to make the non a version yeah non I minutes correct no this this is not comfy it just takes a lot longer and it's a little more complex to playing around with all that stuff Lally but the value is that no I'll tell you the value is I can be in Dubai and a sh can still go out yeah I can be in Antarctica and I can be on break forever second is I can do all sorts of channels I can be the Johnny sense of content I'm just saying right like it's like I expand inventory to whatever number I want and theoretically I'm not I'm not saying I'm going to do this right but theoretically if you trust me and if I build trust with you over a very long-term Horizon then you would trust all the content I put out because right now we do some of it manually we do we have a business Channel called breakdown we have uh case studies and geopolitics and uh stuff in ay we've got Engineering in 100x uh some of these are investments some of these you know we run but the point is why can't I be on all of them people any anyway consumers in multiple forms some people know me for o power some people know me for breakdown you know we did a video on National Logistics policy for example they know me for that okay so the point is this is going to change things and I'm really surprised that companies are not using it but it's not just that it's when you're a video editor you can do a lot more than just create avatars I can take any footage and now I can erase the background so I no longer need green screen so I can be here right now right this video and you can clip me out of the entire background so I can be traveling anywhere I can be doing anything I can actually be doing business and doing some sales or whatever and then just be like okay I quickly shoot something and get out I don't need to carry a green screen I don't need to set it up and pull it up which by the way a lot of creators do so now I can be a Creator you see right it's expanding my ability to be everywhere and do anything and we use a lot of green screen right we need the background to be green screen but we don't need a real green screen anymore because this is getting really good this getting really good and if you're not a fan of paying money to runwayml there are tools like there's there's one called Sam segment anything uh segment anything video where you can run it on your computer and it will remove the background for you great so it's very useful for 3D as well right um we're one of the first few companies AOS which is my parent company it's one of the first few companies to in India at least to do an anime with AI in the US there was a company called cardo crew yes and they made a nice AI animated video we made an animated video too I don't know if you've seen it but I'll show you some snapshots of it so please play it from the starting no I I'll just it will take too long so pretty this thing so I'll just show you some interesting shots of this it's lagging here but other a pretty oh [Music] God so question I really have is an animated video to make a real anime cost you 100K an episode yes now you're able to do a fraction of the cost we we shot this as an iPhone it's a shot with an iPhone and it's got 3D movement because the iPhone has now has an app called cam Trak where you can kind of it uses lar to kind of figure out what the 3D space is like so as I move the camera in 3D space I can make a 3D world that moves along with you this still requires a lot of manual work by the way it is not automated by any means the AI part is like at the end it's like fine it takes the person and it's able to you know make that person AI aish uh but the backgrounds were still generated right in the sense still created in 3D and then you run them through you know your AI workflow it's all IP there at least two three companies we're talking to right now who want us to do an animated video for them saw new Revenue that suddenly opened up and it looks like toys but people are like wait a regular anime costs 100K an episode and you pricing it at $10,000 and you're doing the entire shoot etc etc wow that's a steel so disruption like I said 2024 will be the year of 3D and AI everything from 3D avatars models environments is getting good and there's going to be a lot of hidden AI uh which has been there for a while in tools like blender for example tracking do you know how tracking Works do you know what tracking is 3D tracking yeah okay this this's a technique in 3D called match move and you've seen it probably in every Marvel video okay okay I'll tell you how match move works I'm standing like this I'm going to walk from here to there okay going to walk but the camera is not static it is moving with me so now if I move myself into 3D whatever 3D background I have it needs to move in the same way along with the camera accounting for focal length and all the camera settings technically so match moving is a very very important skill okay and today with the iPhone match moving is very easy there's an app called cam Trak it's free and you can have a match moved character because you you basically get the camera movement in 3D the same camera movement that you did in real life you get in 3D that allows you put put you in any environment so now you don't even need to do that in blender there's you can just choose three four points that don't move or that have moved but are not the subject and it'll automatically tell you how the camera moved it will guess how the camera moved every frame it'll for example take this point right uh let's say this point some point with high contrast and it'll be like did the point go in or out right did the point become larger then well the camera is zooming in the point becom smaller then the camera is zooming out the point you know gets skewed to the left and the camera is turning to the left stuff like that so it's there's a lot of hidden AI there where it's AI but you don't know what's going on and we're actually working on a nice little CGI film right now small one uh hopefully you get to see it soon we're doing it for an intro where we have a nice little bot and by the way these Bots nowaday with Sketch Fab and tools like that they're available you don't even need to create the 3D model in the first place we put Thanos in a video this was not match moved this is just a static shot this was as when you'll see the final video the camera is moving this a nice little robot we have a nostalgic series of videos where you know there's a robot and we're just chilling together eating food in a South Indian restaurant cycling think about it right like there was a time where I really wanted to do this but I didn't have the money or the tools and it was hard 78 years ago do you know how exciting it is that everyone here can now create a Marvel film if you really had just the courage and the time and the ability so for me this is very exciting and that's us doing picking up Trackers in 3D to match move a this thing and we'll put a robot there and the last thing actually not the last thing but one of the most important things which we haven't spoken about yet is images with AI I consider images solved for our use cases for static images in videos to thumbnails banners profile pictures image AI is ready and text in images is the only thing lagging behind as we spoke about but that's not far away it is going to be solved and in fact the way to use it right now is first to generate background images and then to generate the text separately and this this you put in text and it will generate an AI version of the text is solved right there's something called depth mapping that works there's something called control net depth mapping that works really well especially if you want to create 3D text right so we have experimented some of this and is pretty good here's me has different thumbnails as you can see it can modify my body type it can make me look like Superman if it wants to it can also make me look like I don't know who that person is all right so um it's like can you imagine AI is here I mean that looks like a real photo honestly it looks like something from the bloody movie so it's like I'm very excited by the by where it is right now and I'm surprised and worried that not enough people are as excited as me all the me all the media you consume at some point is going to be synthetic completely synthetic in a year or two or at least 90% synthetic do we not see it right and of course 2024 we'll also see AI in things you're going to start putting AI in things in objects we want to focus on entertainment and learning right we also want to put AI in physical objects but we don't want to do it you know the way some of the others are I don't want to make like a robotic assistant that comes and like you know serves you tea or whatever I want to use for entertainment the opportunity for us for me selfishly is to build the next Disney in India and Disney does a lot of Robotics and Hardware but they do it for their rides they have Disneyland they have an avatar have you seen the Avatar the moving Avatar animatronics it's called it's really cool and they have a nice little cute robot uh that they keep displaying again and again it's not just that right like when we create content for example we have these large LED screens we use technology in content already and and for me that is extremely exciting right so uh let me show you a demo of this actually this is an image but um you know we we are doing a small little painting like Harry Potter where you have a painting and you can talk to the painting and it can talk back to you instantly we're starting with chanaka because public domain and chanaka has a lot of insights you can have him in your house like ambient computer all the time and you can talk to him and he can respond and eventually he'll be smarter and smarter because people are building the models and they're getting better at it and we we are synthesizing things right we're putting all this together as a complete package for you and using our distribution to say Hey you know this exists by it so everything we have a face it looks at you there are eyeballs it tracks and hopefully I get to show the next time I'm here maybe I'll get to show some of this to you so we really think that that is still content it's two-way content everything you watch on social media is one way this is two-way but this is not scalable but if you have let's say vun Maya sitting at your house sitting in your house on your wall that you can talk to and you be like what do you think of this problem and it's trained on all the data that I've ever thought about or written and that's why I write so much these days I write like a madman these days cuz I'm like that's the way for me to be immortal technically right to write so that my thoughts my my deep insights are with you and the cool thing is I'm actually a hardware Noob I'm also learning this on the fly right I'm also picking this up on the Fly Shashank from our team is actually very smart at this and that's why I said right the smartest people in India are coming and working with us now because they realize that there's an actual real shot of whatever they build to go out to reach this audience for example so we want to do more and more of this otherwise what's the point of doing anything if you're not taking on easy problems as well as hard problems right we want to do as much as we can in the world of content so I want to tell you some Theses I have okay uh how the world is going to evolve how the world is going to change okay the first thing is I think the world is moving towards outcomes um there was a time where we used to sell software I've sold software in the past I've run SAS companies in the past I've run Marketplace in the past jobs SP recruitment was a Marketplace I've done software for business in the us when we were first Avalon Labs we we WR software for business in the US so I think the world is moving away from buy software at $9 I'll tell you why because any space that's attractive there are lots of players every you know sort of I'm not calling these Founders lazy or anything but they're really good at writing code and it's an introverts thing right like a lot of people have gotten good at writing code in the last few years so they're sitting in their house and they're like I'll make this tool but I'll make it cheap one guy will charge $9 other guy will charge $8 and then some guy will be like this is a big opportunity I'll go raise a $100 million and keep it free so the price to play is $100 million and I don't want to play a game where it's $100 million just to play you don't even not even guaranteed to win I don't want to play that game and I don't want to be free for 10 years because like I said like my friend a says apps are becoming content especially consumer apps one year later one year your app will run after that some new thing will come and people get bored it's like a viral video people get bored more on the next thing unless you have the stamina to keep building apps again and again and again and also getting all of them to go viral again and again and again I really don't think it's a game worth playing okay but people want outcomes the CEO cannot use Ai and do everything he still wants somebody to have context of something so when we sell instead of we can sell a video anime tool and I know other people have done it we saw the opportunity we said bro we have a comfy UI workflow that we worked hard on it's Unique to us we can put this on a server and allow other people to use it in one click but you know what it is better we offer this as a service we'll do the entire thing we'll shoot it we will think of the script we'll think of the idea we'll give you a finished product that's a better offering than us saying use this tool for $9 because next guy will come and charge $8 and it's already started happening and we've learned this lesson the hard way right so for example if you're selling a video if if I'm selling video I'll say bro I I'll I'll do the full service for you why should I do just the tool I will use the tool behind the scenes instead of hiring a thumbnail person instead of hiring 10 thumbnail persons for 10 different channels I'll hire one and make him use the tools I will scale the people but I'll still offer the full service to you look at hair products today I don't know if you've seen hair products evolve have you seen the company Tria you know when they start doing they start they have like a counseling thing it's like an experience it's like a full package right so uh they're saying everything from the doctor's appointment to to how it you know works to counseling on top of it to helping you through the process everything but we'll charge more courses for example there's a time where everyone's like Udi is the best then they said YouTube is the best but you know what we learned about YouTube and courses you should look I have a channel called War level to where I've done a series on Unreal Engine 50 episodes first episode has some thousands of views last episode has 80 views so the hard part about that we learn about education is motivation right in this in this audience most of you are excited but I can still see one person sleeping because life is life is full of distractions right it's very easy to and a course is not like a movie it's like even in a movie today you can't wake up and be like I want to refund in the middle of the movie sucks but you're going in because you're like I'm watching this movie I get a switch off I'm going and watching animal he's going to talk some nonsense he going to beat some people up great I'm going to switch off mentally and enjoy this that's what most people do by the way but the minute you do something like a course it's cognitive effort finish the assignment oh now you have to do work attend the class at this time oh now you have to attend class so it's work and what we have learned is that most students and the reason many of you are here at Master's Union is this is experiential am I right or wrong you you guys went and did the Drop Shipping challenge right so people want outcomes and what we learned is it's also about placements boss if you're doing a course now and you're not doing placements you're wasting time put these people in you can't place everybody but place a good chunk of people because what you're in the business of is delivering outcomes and over time your outcomes catch up to you everyone in ntech has learned this by the way so be outcome focused from day one there's a placement report right by Master's Union you you read the placement report and that's the reason I'm here by the way so many teex keep calling me I'm like we by the way these days when we do brand deals uh or work with Brands we ask for like we do proper due diligence we like send all the material I want to see a placement report I want to see this I want to see that and you can ask we I think we've been a bother to you guys like show the placement report right so our job um as Educators right whoever is going to be an educator in the future some of you will end up being in the space it's now become about outcomes students don't just want content and we know that if you just give them content they will give up after 6 10 classes reliably it's about how do you continue to motivate them how do you give them a peer group how do you give them jobs right so yes you can sell the course for zero YouTube is free but putting everything else together that's what students want and they're willing to pay for it you'll still have to build the product though you still like your hair product still has to work right your video thing still has to work but you may not need to sell the product directly now sell the service sell the solution around it the last thing is we don't use any ai ai in sales zero 0% and I'm very strict with my team that we are not going to use any a AI in sales can you can you tell me why I think when you do sales right it's it's a lot about understanding local nuance and context and the minute somebody talks to bot even they have a slight hinted aot they turn off mentally they turn off they feel tricked they feel cheated a part of your brain is like I got sent a fake it's like sometimes you go you want to meet some government official because you want something done and then they make their assistant meet you you feel cheated right maybe you traveled from Bangalore to Delhi to meet some founder and his assistant met you you feel upset right especially the founder said I will meet you sort of like that they feel like that they feel cheated so we don't use any AI in sales but then I have a rule these days which is we don't have hard and fast rules about anything we don't have hard and fast rules about anything because I know people whove had hard and F like there's some companies who gone out and said we will never run advertising and today they're like oh we should have not said that publicly 50 times right so um smart people trap themselves by taking hard stances on everything and we believe Jeff beos keeps saying this right if you don't change your mind you're an idiot you show static Minds like a static mind set and lot of companies as they get larger do this companies the way I think companies run is it's like you have arteries and sometimes company's arteries get stiff there are blockages cholesterol deposits you get how many of you studied little bit of understand a little bit of how this works okay some of you understand okay your arteries get stiff over time right and uh I one thing I do like I said I spend most of my time creating content but also I'm designing an Aug an organization right and the thing is how do we prevent Our arteries from not being stiff we have rules but we revisit our rules every few months cuz this stuff is changing too fast and I am only as good as fast as I can be and that requires a very different kind of organization requires very different type of career plan how do I tell many of the AI people who work in my team what their career plan is AI and content these careers aren't even there yet there's no name for what many of the people in my team do there's no name especially on the AI agency it's like it's weird it's like we call it applied AI but what is it really so it's all new and weird and there's no career map nobody knows in 10 years where you're going to be so we evolving over time and that's what uh before I end this I want to say one more thing right I think is wiser to look at your career now as a series of projects rather than one career I will eat my hat if 20 years later 20 years later you're doing the exact same thing you set out to do especially in this world that's changing so rapidly like some of you want to be know maybe X or Y or Z especially specific job roles right some some I know a person who was like I want to be product marketing manager I don't know what that is by the way is that product is that marketing I I don't know what that is I'm just being very clear and I've been I've been in the SAS space for a while right um but they're like I yeah I see myself doing this for 20 years I'm like how how do you have the foresight to see 20 years in the future I don't know what's going to happen 20 years in the future right so I think we see a year two years in the future and I think the more you see things like toys but you see toys with like lots of usage like today Apple Vision Pro and we are a lot of people see them as toys I think there's something really serious there and you can start with the most basic of apps you can start with a chess app or something very very simple and today you might not be successful but 6 years down the line 7 years down the line when these things get mainstream there's a small chance you can be successful there's a nice little quote I want to end with right there's a guy called Devon on Twitter he's like Society progresses because there are 100 people out of a thousand people who try some nonsense something new and out of those people 99% 99 people blow up they fail they lose all their money they die they lose their arms trying something new but that one guy where it works everyone copies him and that one guy becomes really wealthy and really influential it's a power law right there's power law in everything so uh so we feel like the world is evolving to a point where everyone needs to keep running experiments not all of the experim experiment will work of course but that one experiment that works you can double click on it and you need to lead your life in a series of projects sometimes serial sometimes parallel they actually find out what works and the entire idea of a career I'm I wrote a book called pajama profit by the way yeah it was about freelance careers sorry we didn't have chat GB back then and I was right about we were right about almost everything all those careers became really popular and successful the book was 2018 they in 200 24 now 6 years it's been go read the book everything is gold everything is still valuable even today but the era of skills we are unsure of what happens to skills in the future and we know we understand that the world is Shifting and there a lot of people who are static whose arteries are stiff and they fight they're like why don't you're wrong I'm like actually I don't mind being wrong but this is what I see and I can't I have to be honest about what I see and what I think and you're free to listen to other people you're free to call me an idiot and say this guy doesn't know anything but at the same time we're working at the edge other people are giving gar at the end of the everyone's giving Gan but we working on the edge of giving Gan right so that might be useful to you and I think what you should take away is what are the things you can learn that will work in any career and all the old things that have lasted 200,000 years are still true today you know what they are ability to sell ability to build relationships ability to think about what's going to be popular a year in from now ability to invest in smart people in smart things that will work a few years from ability to make decisions being kind being nice to other people not screwing other people because you know 10 years later they can become really successful and you know then you know they they attack you back stuff like that it's all the same old rules nothing changed technology didn't change anything just made us who we are awesome that's it for me thanks for listening to me and I always put in a disclaimer always chart your own path don't listen to other people questions yeah shoot I'm what my name is how do you spell it g m okay so 20 21 it's like the blockchain space is quite friendly starts in that space yeah2 233 I guess even this year it's probably going to be AI it will be AI so what is your prediction on the next big piece of technology you know the problem I have and sometimes Engineers also do this is a problem with the way colleg is structured ured civil engineering is different computer science engineering is different every field is different that's what people say I don't see too much of a difference this is my personal opinion because I started with games I mean I games and let's say AI there's so many similarities show there are differences there also lots of similarities right A Game Dev is very capable of building AI models for example um I don't see like to actually build on the blockchain it's not like a decent engineer will be able to pick up solidity in like weeks right so from a building perspective to me it's all the same now it depends on your interest area right and I feel like AI is one of those things where like I showed you it disrupts things in a very weird way it just changes Industries so you still have to choose an industry even in blockchain you still have to choose an industry right there's blockchain for Real Estate by the way which I I with blockchain is very hard to tell which is which is fake which is real right it's the same with AI like I can come and tell you I'm using AI in real estate how much of it is that only when you spend enough time with AI you'll know is this nonsense or not right and VCS have a process called due diligence where they're going through everything that the founder has said and they're like is the technology really real right is this really possible because remember in the early stages you don't know if it's possible you're thinking you're trying your best and you're hopeful and everyone's hoping that in 5 years it become something meaningful uh with the technology risk we don't know if it'll work or not so what I'm trying to say is even 3D you can use 3D in automobiles you can use 3D to make games right you can use 3D to make automobile experiences for example you can use 3D to make environments for movies so they're all different fields know they're all different careers they're all different like Industries so I feel don't get too obsessed with the technology just choose a field instead what are you what field are you excited about I I wasn't blockchain I mean but blockchain is not a field like this is the wrong way to look at it right blockchain AI they're not they are Technologies choose a field like we do AI in content content is the field AI is the tool and I'll tell you one thing I have no need to tell 99% of the people that talk to me that at least on the business side that we are using AI behind this I'm just like we're doing an anime we have a way to do an anime I don't need to explain the AI part same with blockchain most of the most successful blockchain companies don't even need to tell the world that they're a blockchain company it's like yeah I back end me so choose a field anyone else uh yeah so uh uh I'm Anish I'm from Bangalore and I'm currently building something yeah so I'm Anish and I'm from Bangalore and I'm currently building something in the field of Robotics so what I wanted to know is recently lnt published a report saying that there is a like they're struggling to fill 30,000 jobs in the positions of like Carpenters electricians and plumbers so where do you see uh robots that can do specific tasks like carpentry or Plumbing in the future um I just want to say I'm I'm a hardware Noob right so this is out of my depth and expertise I don't do this on a daily basis so these are forward-looking statements um it's very important I wish more more people did this because it at least gives the audience what to expect right so these are my I'm not sure about this but I think in like 2 3 years we will have some level of bots doing simple stuff but I think more than anything what I can see right now is like let's say you want to fix a tap today you would call Urban clap or whatever but I think if you wear the meta glasses and The Meta glasses or let's say The Vision Pro better example and it tells you this is how you fix the tab it's got an AI looking at it Vision model right so got an AI is looking at the tap and saying this is how you fix the tap step one step two step three also we filled your Amazon card with these five things that you need to do that itself changes the world a little bit CU people become more self-sufficient less you know they have to rely on other people less same with assembling computers right uh I don't spend on assembling computers because I assemble computers by myself but I know a lot of people in my team who will go to a store and be like computer assemble cuz Parts you can buy from Amazon you can put it together yourself I enjoy the process of building a PC but I know if you don't know how to build a PC it is actually pretty simple and if there's an AI looking at it can tell you this fits here right and the Vision Pro can show you where it exactly fits um so for me uh that is what what I see first in at least in plumbing and carpentry and stuff like that now with robotics being able to solve that it's at least a few years away and there's like so many things right regulation safety proof that it works the initial versions will be like very expensive everything in the real world takes a lot longer it always takes like decades to play out everything in the digital world happens fast so the the Gap is only widening no the digital world is going faster and faster and faster the offline world is still taking its time um so yeah I think maybe a half a decade a decade till we see some decent progress in India awesome next hi vun I'm bhav uh so you also told that humans are here to stay or there's a lot of job that humans do but recently also put out an insta story telling that cars created as many jobs for horses as AI will create for humans can you shed a bit more light on that yeah so and this is something I I I really want to mention has Nuance in it which is dude if you're top 5 10% of humans you will always have a job you'll always have something to do because what is a job a job is someone paying you money for something somebody wants something done they're coming to you they're saying do this for me see having a salary is like a very disconnected way to make money right especially the father you are from revenue because a sales rep even today is some of their money they make from incentive they have to talk to the customer they to close it and they know understand very well why the customer is paying you the customer wants something you're providing that as a service what I'm trying to say is if you're in the top 5 10% the customers will still come to you because there's something you have that other people don't and you might be using a bunch of AI tools to scale yourself but the problem with that with somebody like me saying I can take on 50 projects instead of two projects because if I was a freelancer now i' would be taking on lots of projects because I can the tools are there and the Arbitrage is high know like I said 15 million people 20 million people who know about mid Journey so the other people are still designing images the old way hiring somebody and saying yeah image B so I would take up far more projects um which leaves less for other people it is as much as people think it's not zero some it is zero some right like uh this is the this is the feedback I wanted to give some of the people in the Drop Shipping challenge yesterday right which is um somebody's making a moisturizer brand I was like do you have any idea how many moisturizer Brands there are on Amazon and as much as you think that oh they've increased the market for me they've not they are it's like now you have to compete with 50 other moisturizer Brands that's tough you're going to get no sales so the first guy is able to absorb far more of consumer demand so that's what's happening with AI that saying is true the person with AI will take your job job AI won't take a job a person using AI will take a job because somebody still has to sell right so the guy who's doing last mile delivery selling taking client projects himself can now scale to 10x and let's say a subpar writer or contractor is no longer going to get projects because projects gravitate towards people who have the expertise the brand and the bandwidth which are all three are increasing with internet and AI so it's more that so no you will not if you're in the top 5 10% you will not use your job lose your job and yes you have to try to get in the top 5 10% you know that's always been true that has always been true the people who are not in the top few percentage of any field have anyway historically don't not done that well and it's not hard to get in the top 5 10% if you know about AI we've done the math right of 8 billion people if you know about AI you're already there because you don't understand what 5% of 8 billion is what is the number what is 5% of 8 billion about 400 million 400 million Char has 200 million users so you're still safe you're still fine you're already in the top 5% of your field if you're able to use these tools so I want you to understand that the population of the world is very very large especially in India it's huge it's huge and the fact that you're sitting here means you already have some amount of money at least through through your parents you have some advantage and resources that are anyway putting you beyond so many more people in India India is a tough place man most people forget about chat GP they don't have water they don't have food so you remember when you talk about top 5% you're also competing with them you'll be fine don't don't stress about it and just don't get too married to a career cuz one thing that will happen is there'll be some shuffling of careers and um what is commercially viable and you'll shift you can't expect like a this is a problem I keep software Engineers keep tagging and irritating me sometimes right and like I I wish I could tell them this I wish you understood the Nuance in what I'm saying if you're in the top 5 10% nothing's going to happen to you nothing it's not going to affect you cuz you'll shift to a more lucrative software engineering job you can move to games you can move to cryptography you can build an OS tomorrow if necessary right there are so many opportunities there harder problems or different problems why are you married to being the next front-end developer everyone wants to be a front end developer yes sure it's cool people have made it cool there's lot lots of YouTube content on it but by the time it's become that cool it's already oversaturated no so Supply demand rules apply everywhere so why you have an very useful skill why go to the one place where there's a q look people do this right they go to a bank and there's a queue they like let's say five lines and there's a queue you will not go to the line with the que you will step to the next line with there's two people but here this is the one place where everyone's like okay Q is very long but I'll Stand in the queue the same thing I see with like all the competitive exams right and I till today I don't understand it but I feel that's because of limited exposure and knowledge of what other careers are available and one of the things I hope to do is there are lots of alternative careers that where people are doing really well like this is an argument I keep having with UI designers right um You all want to be designers and you're all learning this useful skill but you all want to go work for cred you all want to work for the same bloody purple crypto website you all want to make the same purple crypto website but go to Real Estate the real estate guys all the large real estate brands are making insane amount of money and insanely profitable and many of them public companies they really need help with design because if they sell a property they make maybe a CR on it that's the margins are insane right because they bought that land so many years ago sometimes they're just selling land and it's about making the land look cool go design brochures for them go design websites for them there's so much money to be made there but no I want to go work at cred why it's cool I've seen memes about it on Twitter so uh the group think and consensus in India is very very high and one thing I've noticed is the minute every time I try to break group think in consensus I get attacked I'm like do you not see that you are the mistake right do you not see that group think is the problem there are opportunities just disperse little bit and all of you can make insane amount of money in these different spaces but if all of you stack up in front of cred doors then some of you are going to get opportuni some of you are not it becomes like a rationing system Right comes like an competitive exam all over again and this fascination with entering a queue or a line where there lots of people is I don't understand it in India I don't get it there lots of careers and it's all sometimes it's about talking to to other people and be like okay where is the extra money being made right and moving there think of your career in terms of projects rather than one static career and your life will be so much better you will have much less anxiety you have anxiety because you're married to whatever you studied and you're like this is the thing I'm going to do hey Von I think sitting in this session today was my unfair advantage and apart from that on your note of outcomes I've come to believe that outcomes as as how you put it is more of a service and I've stuck by the Nuance that at least Building Products is where the scale is and service is something you shouldn't get into and in type of an agency would okay I will I will break that down for you right I'm not the biggest fan of agencies right I I I really hate the idea of agencies because it's a people problem okay or at least in my head I've like it I think it's a people problem but let me tell you some uh learnings that I've had right what what are the largest software companies in India TC associated with software software scalable tell me TCS TCS ppro infosis come on more HCL HCL by market cap also they're pretty large they do software how many of them have successful products how much of it is Services how much of it is product all of it's almost all services right so you're telling me that after all this drum beating about software this services companies the biggest software companies in India services companies now let's look at the actual SAS companies there's some in India right like for example tally is doing well a bunch of others but double click on their businesses and you'll realize a lot of it is Services they're cleverly disguised Services the products are good first foot in the door but people really want stuff done in India and they don't mind spending some extra money when we were running scenes right A lot of people came to us and says in just giving us Community software run my community very hard to do by the way run somebody else's Community very hard to train for it's very random you need distribution it's it's pretty hard but that's what people really wanted if if I had successfully run communities for other people and done that at scale you would have probably made a lot more money because that's what people wanted so and even Zoho right we look at software as a service SAS successes in India and we say look at Zoho this that dud most of Zoo's business comes from outside of India Shar vmu has been outside of India for so long no decades so you need to cut through the noise to find out exactly where is the dollar coming from exactly who is paying that's my job right exactly where is the money coming from and one thing I'm realizing is this pure I will pay $100 a month is not is not as valuable as people thought it was like what is zomato today can run even if people pick up the phone and dial a Central service and the service says sir what do you want to order and it says I say pan but Masala and they say okay here are the three restaurants and I pick one they still work but the app allows them to scale faster it's a distribution mechanism where you know you don't need to do the call therefore you can onboard more users it's a place to send all your users at once but what their real MO is is the logistics Network can we agree because app many people can build food apps and there were many food apps but the ones with the largest networks the ones with the largest competitive Advantage is one so uh I'm a Believer at least in India and this is very nuanced to India because in the US people value their time so they buy software at least in India we don't value our time enough businesses don't value time enough and they will always choose to hire a person and dude I've seen it in my own behavior we will always choose the cheapest plan of a CRM it'll always be like I'd rather hire a person to do this than it's like it's it's cheap it's actually cheaper right and um so I have found when double clicking that it's a good it's good in theory if you have a product that scales like crazy but if you have a product that's also competition and competition does price cutting and people really want outcomes the customers really want outcomes they will eventually gravitate from that product to somebody who says I will do the entire outcome for you so you'll anyway lose your contract to the outcome but see this requires a different mindset the VC mindset in India is you need a product or the VC mindset in general is you need a product and that's where the scale is going to come from that's how how you're going to get big but I'll tell you one thing VC have changed their mind at least in India have changed their mind on this because they have not seing successes no once you start seeing successes you're like success is still in the services so how I've changed as a human being about this is we will use products internally we will sell it as a service and we will hire really good talent and maybe I'm not the best at like managing client expectations and all that we'll hire for it and I'm learning so much about organ organization design in fact this year for me it's less about AI right because last year we did a lot of that this year is like how do I build an Aug that outlasts me and I'm learning so much about building organ I can talk about for two hours about how to build an organization all the things that I've learned right uh in fact I I think the organization is a lot like the human body you need to have an immune system for example right if if if uh when you start competing with other people you sort of like get into a fever you start consuming more glucose more oxygen like there's lots of things that goes on running an A so it's more about that and if I can do that right then you can scale anything no then you can scale Services you can scale all of it and I don't want to build a VC scale business first get to few crores in Revenue then think about how do I build a, CR Revenue business awesome yeah uh sir hello oh hi uh my name is Aur and after listening from from the starting I just had one question that I would ask you in in my mind but when you started speaking I just wanted to ask you question collectively from our batch see we as a batch we are making a lot of startups right but when the when it starts it gets a bit tough right and many of us just give up at the end when I saw you right now it's like what keeps you going is it your just sheer curiosity because I follow every bit of your content whenever I see you speak you there's always like bit of curiosity in there so what are some things that help you drive yourself to achieve everything because your portfolio is are so diverse and what are some work ethics that we can take from your career to get success in our entreprene Journeys just a personal answer right I think once you get into the top like the top Eon I'm pretty sure I'm pronouncing it wrong but when once you get in the top strata of of what you're doing and you're known for it you don't want to fall from grace like you don't want people to think oh he was he's an hasb been you don't want to be a caution story right um and also it's like we're good at this so it's very easy for me to slip and give the throne to somebody else right I'm not going to I'm not saying we have a throne or anything I'm just saying like you're you're good at something might as well do it see there's a very interesting story about Enlightenment I told it to some people yesterday I'll tell it to you again I keep telling the story in my team okay I was the I'm the most experimental person you'll meet I'll keep trying stuff back to back to back l lots of failures in life simply because I love radically experimenting last 8 months I've found a new strategy which is I will experiment but I will partner with excellent CEOs who understand the value of pmf cuz I don't care about pmf as much I'm like because I found pmf on multiple things in the last 6 months alone right uh I I'm I'm pretty sure I'll find pmf again on some new thing right because I'm I I'm trying to look for the truth truth seeking the Insight in everything and um when you're like that and it like I'm not thinking from a place of ego I just really wanted to work then you find pmf right um so I found that I'm not the best at like sustaining one thing but I keep telling the story to my team there's a lumberjack and I'm butchering the story but I'll tell you anyway there was a lumberjack and he was very good at cutting trees and what he used to do on a daily basis was he used to cut trees used to chop wood and then he used to carry water to carry water from one place to from the well to his house or whatever then one day he said I'm going to be a monk okay and he gave up everything he decides to start meditating then he becomes enlightened he reaches Nirvana he's like now I know everything about the world I know everything and then he then people asked him now that you know everything now that you're enlightened what are you going to do on a daily basis he says I'm going to chop W carry water and that's what I've learned right there is nothing once you start making money there is there is nothing Beyond this there is nothing there is no reason there's no there's no reason for me to make one more rupee I'll be happy I'll have to live within my means but I'll be happy but then what do you do what do you do there is nothing to do so we chop wood and carry water right and I've been thinking a lot about this right do I want to build a legacy what do I leave this for why should I make the next why should I make the next rupe and these are struggles where to be very honest I have not figured it out but I don't think there's any human who has and and there's some like stretch thoughts right like okay what if I can be immortal and all that but the technology is not there yet um so I think I like what I do because it's the only thing I know and therefore I keep doing it when I was very young I had distractions I was 2021 I was like I was young right like I can do this I can do this I can do this today we don't have distraction I've tried everything cool thing is in 10 years I've tried everything that I love and I still keep trying new if I loved it I have tried it but then you learn that not everything works and the things that work require chop chopping wood carrying water so that's why I shop would carry water that's why even though I've made a th000 videos I'll make the Thousand first but in one of your podcast with Nan Kar you said that you don't really like coming in front of the camera and that's why behind think you want to automate everything yeah do you think chopping wood carrying water is fun after you chopped to carry water carried water for years and decades no it's not it's not fun I don't see it as fun and that's me learning and evolving right it's not fun therefore I will use technology to make it try to make it as fun as I possibly can but the day I decide it's not fun and I retire there will be somebody else who does what I do and they will go through the same conundrum thank you sir yeah this is m fromb b I wanted to ask you talked about creating organizations like big organizations so as a first time founder who worked in his family business come from very supportive background doesn't have the freedom as such so I I really struggle in managing teams but I have the idea I know I can do it but how to build that on a truly scalable level how I can what is your how old are you it's very hard for somebody who's like 24 25 to to manage a large how big is the AUG yeah it's it's hard no because it's a respect thing um Henry Ford uh he was known for his genius of building the car uh but he was a bad manager he when people made alternative cars or whatever new models he he tore them down and he passed that down along his family so two generations later the next Ford he also like horrible towards his managers because he was not getting respect from them so he tore all of them down and I think step one at least with family businesses if you're the son of a leader right like sometimes I think about this I have a team and suppose I had a kid I don't have a kid but suppose I have a kid someday um is my team going to feel threatened by that person because the team would have worked there for 10 years and this kid has just walked into something that works how does that play out whose ego gets damaged do they understand that the that the kid is the you know on a pedestal why we a mer meritocracy everything that everyone's accomplished here has been accomplished because they put in the effort in the time and the energy why is my kid special so it's that it's earning respect and I feel like that takes till you are 30 40 it's one of those things that you can't easily codify right like respect for your team is important in me increasing Top Line but yeah soft soft skills so how do you gain the respect of your team my me would be like showing up that's not enough that's not enough you need wins you need to do something radical and needs to work you need to be misunderstood that's Sam Alman thing right you need to be willing to be Mis understood in the short term and then something works later you need that uh which can also blow up in your face you do something stupid and it doesn't work out then they think this guy is an idiot but that's the thing know even with tribes you had this when you reach 18 years old in a tribe You' go out and you have to bring a lion's head back to prove that you know what you're prove that you're good you're capable of leading the crowd very tough very tough um and that's why family businesses ruin themselves after two three generations it's mostly because of these soft side things versus the hard side cuz strong people usually create the next next generation of really weak people cuz you've not gone out and like hunted yourself and come back you've always been hunted for someone's always given you food on the table um and the problem with that is not somebody's giving you food on the table the problem is the team has seen that you have been given food on the table for number of years and the team is like and I've seen that at least in families eat you know people up um I don't have a solution I've never faced this problem but this is just what I've seen from the side looking at some family businesses we take the last question last two questions yeah um my name is I'm also part of the YouTube b um uh you you into building games and you also want to games in the have you seen the movie yeah so like V glass which you can see through ultimately and only directional threadmill Disney has built that only direction threadmill or tiles I would say and they're not good I mean it's got iner It's got inertia problems but yeah at some point it'll be solved yeah yeah like and Vision Pro is technically also almost there to solve the same thing and since you want to incorporate Ai and yesterday at my install you told me about building the low and that's what I think you're doing right now uh and to benefits later on so how do you uh like what's the idea of incorporating Ai and let's say building a game like Ready Player One in future with your brand yeah I'd love to I I really like this game called rust uh which is like a survival game right so I love rust I think there should be a rust in VR uh but I'll tell you I'll tell why we haven't built it yet we've built a card game and stuff right uh not enough people in India have VR headsets our distribution is very India focused so if I build a game it like a VR game and not a mobile game it doesn't take advantage of our distribution which means I'm starting from scratch because I have to build Us distribution for the user base how much of rust Revenue comes from India very little right how much of World of War crafts Revenue comes from India very little so first let me really take advantage of the India distribution right let me build things for India let me solve that and then I will take many of these riskier bets uh we built a game it's called op out the card game last year we built uh I'm in the game you can play me as a card right um so is all the cool people in our company you can play them as cards um uh but we want to build for mobile because it's like who has people don't even have desktops in India I mean um like building a game like one would be like a breakthrough industry and but why not just Target the global market and just I don't have the distribution no it's a much more expensive business when I don't have distribution yeah yeah let's let's let's do some a different side of the audience sure thank you for the this is a personal problem that I faced is there are like hundreds of tools of AI out there and every tool there is a learning curve and if you are learning a tool right now there might be some other tools that might come up which is better than how do you actually discover as well as you know learn that tool and actually apply start applying in your business follow at the r theun Maya on Instagram and overpower because this is my job right like it's kind of like this is exactly what I keep doing like I'll give you an example the way it's moving IP adapter face was great for thumbnails then we switched to Insight ID uh sorry uh instant ID which is like it took a week to switch and we were like okay this is better and we tried Photo Maker also in between which wasn't great everything is hyped and my job is to sort of figure out out of the hype which is the right which which is useful and everyone knows we do content so if I find it useful like all my thumbnails on my last few videos are all AI so we're using in instant ID now but for the rabbit R1 video we used IP adapter face we knew that this is going to get flooded with tools we knew it's going to get flooded with open source paid etc etc right and that's exactly why we created the channels we like demystified we knew there's going to be infinite content buckets here so just stay up to date like hit the hit the post notifications button or whatever I'm serious it's it's we do this because we think it's how much money do I charge you to watch my some is on Reddit okay somebody was making some AI course or something and somebody in the comment I really love that comment like you don't need all this just follow War Maya for free on Instagram I've learn more from a shots seriously shots and reals right seriously I mean I just want to demystify it for free like No Agenda so with this you pre distribution yes so with this you prefer to go with monthly plans for any AI you know I'm a believer that once you have a GPU it's all free if you're smart enough once you have a GPU it's all free like a good GPU is like maybe a lak today uh but that's it you now have to pay for another subscription again just electricity which is cheap GP is a graphic card with comfy UI every tool will be [Applause] available thank you so much I hope this is avable

Channel: Varun Mayya

Views: 162,409

Keywords:

Id: JiZqV1CSfMY

Length: 108min 22sec (6502 seconds)

Published: Sun Feb 18 2024