AI Innovation in India

For our next very special, I would like to call upon three very special young innovation champions on the stage and share their experience. We have with us Srinidhi Bagla, Jai Vardhan and Adhiraj. Please come on the stage and share your journey. Thank you.

Hello, my name is Adhiraj Chauhan. And I’m a high school student of 11th grade. And I’m the founder and CEO of Delta AI Revolution, Delta standing for change. The reason my company is called Delta AI Revolution is because I’m a very, very passionate entrepreneur who believes in the intersection of solving societal issues with modern day technology. So I firstly like to extend my heartiest thanks to the Atil Innovation Mission. It is in their Atil Innovation Tinkering Lab, which I started my project and created my first MVP. Also to Intel for providing support. It’s important mentorship and to my very own school who’s provided. We support and been there with me every step. So my journey started when I realized that amongst the youth in our country, mental health is an epidemic.

And despite a lot of efforts because of a large population, the ratio of psychiatrists to people is one psychiatrist for 100 ,000 people. So my startup is a mental health support platform. It is an AI -driven platform training different therapy techniques ready to cater up to more than 100 disorders. We provide our platform to different psychiatrists firms such as Dr. Mora Psychiatric Clinic. And we are also in talks with the Delhi Psychiatrist Association. We provide our platform to them which they can provide to their clients. We’ve touched over almost 20 clients right now. We’re shifting to a B2C model. I’d also like to thank the Ministry of Electronics and IT who has provided me funding. And again, I’d like to thank Agile Innovation Mission and Intel for providing me this opportunity as a young innovation leader and a young entrepreneur.

Thank you so much.

Thank you. Shreenidhi please come on stage and share your experience.

Hello everyone myself Shreenidhi from BG’s National Public School Bangalore. I’m very grateful for everyone who’s been part of organizing the summit for giving us this wonderful opportunity of being here and presenting our project. It gives us confidence to build something new and gives us confidence that people believe in the youth today and innovation just doesn’t depend on age it depends on intent. So my project is basically charades named after a game which most of us might be knowing dumb charades where where the players are supposed to explain a movie or a song name without using speech and only hand. I decided to name my project charades because this is because the game is similar to something similar to what we try to help.

It is a glove that converts sign language to speech and speech to Braille trying to help the deafblind community. Right now we have developed our models over thousands of images using machine learning deep learning and all of this was possible only because of the boot camps from Tinkerpreneur Challenge, the mentorship programs from Atul Tinkerpreneur and Intel, Neeti Aayog, the mentoring sessions held by the summit organizers and we’re really thankful for everyone who has been part of this summit. Yeah that is everything I would like to say right now. Thank you.

We have our next next innovator and I don’t want to introduce him he’ll introduce himself and it’s going to be a very surprising and his journey is very surprising and let me call him on stage

thank you ma ‘am and hello everyone I am Jaywardhan Tyagi so if I’m a bit clean I just recently got appeared on Shark Tank India where I secured funding from Sir Raman Gupta founder of Boat Lifestyle and a founder fellowship from Sir Ritesh Agarwal who is the founder of Oyer Rooms so yeah to start with like let’s describe myself on broader spectrum I am an engineer I’m a student and I am a reader so so so broader AI in healthcare has evolved structurally over the recent decade. Like, if I had to describe radiology AI in 2016, it would be like a metal detector at an airport. But today it’s like a full airport security system with a CT scanner, with behavioral analytics and security cameras and, you know, all.

So, we have seen amazing benchmarks, especially from University of Florida recently, this year and the previous year’s end. And we have seen great progress in medical vision language models. But the question that matters isn’t how well these models perform on these curated benchmarks. It is, will they maintain this performance when the distribution shift is introduced? So, the distribution shift is like some edge cases, which are not so substantial. Like, if we talk about radiology, an input from a newly installed MRI, with a different contrast that can be considered uh considered as a distribution shift actually vision language models today uh are very poor on handling those distribution shifts they hallucinate a lot so basically uh the problem isn’t the architecture itself but it’s the thinking that okay a single model has the power to understand like every part of every dynamic of human health which is of course possible and you know but this is less than a technical necessity and more like uh you know obsession with scaling so yeah so basically um what we have derived it’s it’s almost like thinking a transcription model which doesn’t take audio as an input but takes video frames and just just try to determine what the person is saying from those videos it’s possible but inefficient so what’s the solution there The solution is a system or a framework that reasons across modalities and refers to previous conclusions, contradicts them, and finally describes them all in an understandable manner rather than a clinical report.

So, yeah, it turns out I’m working on the same thing. So, yeah, so before I describe Neuropex EIS technology, as it appeared on Shaktang, it’s good to first clarify what it is not. So it’s not a classifier for, you know, narrow disease prediction tasks. It’s not a standalone VLM with a reporting layer attached, and it’s not an orchestration on a GPT. So now let’s, like, discuss what it is really. So we have two pipelines. One is for radiology, and one is for dermatology. A radiology pipeline has dyno plus clip plus retrieval augmented vision language models, which actually… are able to understand multiple sequences of MRIs and can read the x -rays as well and can describe them in real time using clinical language.

It’s still in the active development when it comes to structuring those findings, but yeah, it’s still in the game. So the older radiology pipeline, which near the shark tank time, that was like a segmentation model, which took in 3D MRI files and just segmented the three tissues, CSF, gray matter, and white matter tissues in the brain. So what happens is when you have those tissue segmentations and you have those proportions, you can actually risk for a wide variety of neurological disorders. That was all of the radiology pipeline. I plan to actually show the demo as well, but we have time constraint. Yeah. So. Let’s talk about deep down then. Deeddom has a visual language model that’s trained on Demoscopy, Clinical and Histopathology Datasets and So you first describe your problem Vocally and then you answer A clarification question And then it just generates a report And it’s live out there, you can just sign up on the Neuropexia Site.

So let’s cover up It seems no less than a mission And this mission aligns with India’s goals of Leveraging technology for an outcome Driven impact. And yeah, it turns out We’ll be working on it So yeah, thank

Thank you so much I would now like to invite our mission director Atal Innovation Mission to have a few words And address the audience

Thank you Thank you Thank you Thanks Arunima Such a pleasure seeing you all here So many partners You know it’s amazing Were you guys listening to what they were saying These kids It’s unbelievable You know I just finished A session, this was on the future of work And I was coming, there were four of us And I was telling them the biggest challenge For us will be The first is I asked them to raise hands Of how many people have been laid off There was only one And I told them He’s the only person ready for the next 10 years It’s very important And you know the problem We are trying to solve on mental health That is going to be the biggest challenge Going forward The disruption is so immense That the ability To re -skill and re -do ourselves Is going to be so high And it’s going to be generational So I think people who have just gone into the workforce And at least for the next 10 odd years Otherwise which are going to face the brunt of it.

And that’s where things like this are going to be critical. But what I was saying there is, and which is going to happen here, in the next 96 hours, Sarah, you and I will celebrate our 10th year of the journey. But more importantly, we will also celebrate the 10th birthday of the Atal Innovation Mission. And just imagine, it is a 10 -year -old, which is today the world’s largest grassroots innovation mission. It’s unbelievable. And this is where you’re seeing what is happening. See the results. These are the ones which are now just going to take on that new India. And that is what I was saying there, that the big challenge is not going to be creating jobs, because just now, we are looking for 1 million jobs a month, right?

So far. Now we will have 12 and 13 and 14 -year -olds ready to take on tasks. We are fossilized completely. And the point here remains that that is where I say two points. The biggest delta multiplier of AI, the benefactor of this is India. The biggest benefactor of AI as a delta multiplier is India. I’ll tell you why. 1 .4 billion will be 1 .6 by 2060. 1 .6 billion people completely empowered. And starting from a low income to shoot up to be one of the biggest economies of the planet. You see the delta? We finally have a delta. We have a tool which is going to make that happen. thing is, for some of us, ma ‘am, we might actually see it happen in our own lifetime.

It is going to be so fast. It is so rapid. And the biggest benefit there which comes is two things about India, which are our biggest strengths. Think about it. The first is the ability to work in an unstructured environment without a playbook. You showed it. The way worst example in human history which happened, the biggest calamity was COVID, the pandemic. There was no playbook. You did not know what to do with it. You emerged as the strongest economy within COVID. You did it. It was unstructured. No one in the world had a playbook. The biggest strength of all of you, and we look up to you as the future which you are, and you’re going to be creating.

the superpower of the world, the biggest strength of India is getting a job done, regardless of the resources available. Ask an Indian, he will get the job done. And Sarah, that is what is the strength of this Jagannath. Guys, today it is about you. Really fantastic. And you know, we are so lucky we have our old partners with us, who started with us right away. Thank you, ma ‘am. Thank you, right from the beginning. You, Sarah. Thank you for walking this journey with us. It’s a long way to go. We have a lot to do. And we are all with you and behind you. And actually looking forward and looking up to all of you.

So thank you for making us proud. Very well done. And your presentation? remarkable. Thank you. Thank you very much.

Thank you so much, sir. Now we have with us a very special guest, Mrs. Sarah Kemp. She is the Vice President International Government Affairs, Intel. I would request,

Good afternoon. It’s not very often in your life that you get an opportunity to make such a difference. And so I want to start by saying thank you, because this journey of 10 years has been life -changing to all of us. And I want to start also by asking all of our technologists, to start… our future technologists, to stand up so that we can properly thank you. So all of the future technologists in the audience, I see you all with your – Stand up. Thank you. You are inspirational, and you are what gives me hope for the future. When I read the headlines and I get a little pressed, I take out my Changemaker brochure that has all of your projects in it, and I think, wow, there is hope for the future.

And I would also echo, I think India’s superpower is absolutely its people, and it is what’s going to make a difference. And I also want to say that I am so grateful to the Indian government for their support. For how they have – teed up and how they are framing AI. At this summit, not only is this summit making history because it is the first summit in the global south and it’s going to lead the global south and India is going to lead that, but what I’m really excited about is the heart and the human that India has put at the center of AI and making sure that the AI is to help people first and foremost.

And so to our future technologists, we put on you a great responsibility because with great talent comes great responsibility. You are looking at the future and you are looking at the future and you are looking at leading us forward. You have the ability to make the society you want, to make us a better version of ourselves by using AI. for good. And I just want to say I am very excited because I have great ambitions for all of you. But with that, I do want to just thank all the partners and look forward to another 10 years. And before we know it, we’ll be there. And I just want to say again, thank you. On behalf of Intel, it has been an incredible honor to be able to be a small player in this.

So, thank you.

Thank you so much, ma ‘am. That was really inspiring. I would also like to mention that, you know, we have top 50 students present, you know, AI tinkerpreneurs present with us. And they were shortlisted by Intel and Atal Innovation Mission by rigorous evaluation. And they were trained and, you know, mentor session was done. So, I would request the dignitaries on the stage to unveil the tinkerpreneur compendium. ma ‘am sir can you please unveil the tinkerpreneur compendium yes we can also have the three young innovation champions jaywardhan srinidhi adhiraj to come can we also have hufeza salim yes on the count of three you can open the ribbon three I see very less energy, you know. Thank you.

Thank you so much. We actually have… So, you know, as our mission director just said that this is the 10th year of Atal Innovation Mission, I mean, like, we, everyone here should be very excited about it because something that you’re seeing right now is being seen only by you. Nobody here has witnessed… the logo of 10 years of Atal Innovation Mission. What they are holding in their hands is the logo of 10 years of Atal Innovation Mission. Can we have a huge round of applause from the crowd? We are also going to play a video. We are also going to play a video. Thank you so much, sir, for joining us. Okay, let’s move on to our next session.

We have a very special address by Mr. Ojasthi Babbar. Can you please come on stage and

identify whether each one of the AI innovations which are happening all across are actually worth backing or not. Otherwise, it’s all noise, all hype, and we try to stay distant from them as such. But having said that, this is the framework for our evaluation. What exactly do we do? Once somebody passes on this, off with this captive network framework, how exactly do we help? How exactly does one incubate, accelerate, and invest? And what kind of value addition do we bring in while we have spoken about them bringing in that kind of value? The first one is rapid validation, if you move on to the next slide. The incubator, the accelerator, and as an investor, we help in rapid validation of these ideas.

The earlier side, though, but we can probably lock on to this slide as well. So we help you stress test that particular feasibility. We help you stress test whether your particular solution would actually work in the real work or not. By bringing in the right corporate client, by bringing in the right pilot partners as such and making sure that the rap… So at the incubator and at accelerator, we have a philosophy. We say we need to fail fast, but we need to fail forward. We need to learn quickly, iterate quickly, and move fast. The second one is, of course, of the controlled pilots that we bring in through our corporate partners. We have a corporate adoption program which we utilize wherein a lot of corporate partners plug into the incubator to give in problem statements which are solved by different entrepreneurs at each one of the different levels.

So that is one of the other programs that we have. Post that, there’s a litmus test that we do and that we help out with is by making sure that there is the right revenue model associated with each one of the startups that actually present and that are actually incubated as such. And here in terms of… These revenue models, we help them optimize the inference cost. I think we’re short of time so the essence is to ensure that we make sure that there’s enough revenue which is coming in, the revenue model is right and tight and that can move forward and get to a global scale level as such and of course the last one being making sure that once you’re growing you would be in need of capital and that capital comes in with the right partners, the right strategic investors and the other stakeholders as such.

Stakeholders like Atal Innovation Mission, like Intel would probably play a very important role when you’re scaling up from 0 to 11. So moving forward that’s the last slide that we have that is actually the gist of our entire AI thesis as such. We believe that any AI innovations which would actually thrive in an Indian ecosystem would have domain depth they would have the right proprietary data which they would utilize to create a mode, a barrier to entry as such and given multiple returns and relevant returns as such and of course having infrastructure railroads like that we have in the country as such if they can utilize the distribution access that we have I think we have a winning equation right in our hands for all AI innovations all across.

In the interest of time I’ll just stop up there. Thank you.

Thank you so much sir. I would now request Ms. Sara to please felicitate Mr. Rojasvi. Can you please come on stage sir? Can we have a round of applause? Can we also have Adhiraj and Jaywardhan to come on stage? We would like to honor you with something for being such good innovation champions. Sara ma ‘am if you could do the honors. Thank you so much. We now have our next speaker. He is the founder, co -founder and CEO of Hooper AI, Mr. Gaurav Dagongar. Can we have a huge round

Since I know we’re pressed for time, I’ll get going right away. I must say I was extremely happy today to come here to get a chance to talk about Hooper. But I think what’s made me really happy is sitting right in between Jayavadar and Srinidhi. I don’t think I’ve felt that energized. I’ve felt energized in a long, long time. Since we are a music technology company, let’s do this a little differently. How many of you recognize this tune? You get it, right? Thank you. Had to. This song released 50 years ago, more than 50 years ago, composed by R .D. Burman, written by Anand Bakshi, sung by the great Kishore Kumar. In 2016, and the reason I had to bring this up is I happened to make a cover version of this song that became really popular.

A few years later, a mint brand launched in India using this cover version as their audio campaign. And as I checked last week, over 100 startups have still used this in the last three months. To promote their product or their brand. Now the question is, have they got a license? Did Anand Bakshi get paid? Did R .D. Burman get paid? A little selfish, did I get paid? A lot of youngsters here who will make covers or who will make originals in the future need to ask this question. And that’s what we do. I’m Gaurav Dagaonkar. I’m the co -founder and CEO of Hooper. And I’ve made my passion my profession. I graduated from IIM Ahmedabad and became a music director.

So for a long time, I made music for films. I’ve had the fortune of having folks like Arijit Singh, Sonu Nigam, Shreya Ghoshal sing my songs. But after 10 years in the music industry, what I felt was, one, India loves its music. Whether it’s our films, whether it’s TV, whether it’s ads, all the deals, 6 million reels we consume daily, they run on music. And yet, when it comes to music rights and music licensing, there seems to be no knowledge. That’s an opaque space. I bet if there’s any entrepreneur in this room, who is using a Bollywood song, do you know how many licenses you need? The better question is, I don’t think, did you even know you needed a license in order to use it, right?

And that’s what we’re solving. Before we built Hooper, India did not have a single platform, even one, that could actually license music. It gives me great pride to say that Hooper is India’s first native, homegrown music licensing platform. And of course, we are a part of the Atal Innovation Mission ecosystem, so that makes me extremely happy. In a nutshell, we are a marketplace, where on one side, the largest labels, the largest artists come and list their songs. So you have folks like Yash Raj Films, Universal Music, even people like A .R. Rahman, next week we’ll get Hanuman Kind, listing their songs. And on the other side, it’s basically brands who come and like it. the music.

Over the last couple of years, we now have over 3 lakh of India’s biggest influencers and 220 brands that are licensing music from us. And it works in a very, very simple manner where the song gets uploaded on the platform, a brand discovers it, licenses it, and the royalty or the revenue goes to the artist. Beneath all of this is our AI infrastructure layer. And it works in a really cool manner. First, when a song comes in, we process that raw audio. We use a multimodal AI there to create different tags such as mood. Is this a song? Is the song a happy song, a sad song? Will it go for a fashion brand or a sports brand?

We also use LLMs to understand brands and try to create some kind of a fingerprint for every brand. And then we try and match the two. What music would work, say, for Baskin and Robbins? What music would work for a Dairy Day? What music would work for a Baskin and Robbins? What music would work for a Baskin and Robbins? mantra and so on that’s essentially what we have done and now it gets exciting because we’ve legally licensed music from authors and composers we can now build on top of it what if say for Mahendra Thar I want to create a hip -hop remix and I want to do it legally and ethically so that the artist gets paid and I think that is where I would love to invite many of you who probably have music as a passion and would want to build on top of the Hooper stack that’s a bit on our AI layer I love doing you know I love my job because on one side we’ve got the largest creators using the platform be it folks like Ranveer Brar, Ashish Vidyarthi, Sadhguru, the Chief Minister of Maharashtra Mr.

Devendra Fadnavis’s YouTube channel uses Hooper and I hope that this year we also get a chance to soundtrack our Honorable Prime Minister’s social media content and videos and apart from that we also have brands large brands like Himalaya, Myntra, Mariko as well as startups that use the platform I’ll just take half a minute to play you a short audio visual that will give you a glimpse of what Hooper has done in the Indian soundtracking ecosystem If the visual doesn’t load I believe it might be better I just want to sing a song That’s so good, that is nice It’s pigeon India Oh Thank you. Thank you. the AIM ecosystem in trying to ensure that India tells better stories, tells them legally, ethically and responsibly.

Thank you so much.

Thank you so much, sir. So we would like to felicitate you if Saramam could do the honors again. He played some music. At least we can clap. So today we have with us top 50 AI thinkpreneurs. You know, these are the people who got selected from about 3 ,500 applications and they are here today from each and every corner of the world representing at the AI Impact Summit. Before I call them, to stage, to give them certificates. I would like to request Ms. Dipali Upadhyaya, our program lead, Ms. Sufeza Salim, Mr. Sumit, our admin and finance head, to please felicitate Ms. Sara Kemp, who is, you know, a huge partner. Intel has been supporting us in training, mentoring, and the selection process of these top tinkerpreneurs.

Thank you so much, ma ‘am. Thank you. give a chair. Everybody please give a chair. These are our star students and you know yeah so Shubham is here to you know felicitate them.

Yeah so from DAV Centenary Schools do we have? Yeah please come forward and then from Infant Jesus School Infant Jesus yeah and ML Khanna the mentors, the teachers as well as the students. Come forward please for a quick photograph. Just come forward please. Yeah Take your certificates and stand You just hold the certificates and take a picture and then why don’t you also come Come come come Come in come in Come in come in Come in come in Come in come in Come in come in Then we have Vidyashil Pagadmi, Radiant International School, Lakeford School and KVIISC. Please come forward quickly. Vidyashil Pagadmi, Radiant, Lakeford and KVIISC. Silver Oaks, Silver Oaks, JSS Matriculation. You can also come forward please.

Join them, join them please. Go ahead. them. Yes, please. The next lot can come. Yes, please. Yes. Somalwar School, Father Eggnall, Murarji Desai, please come forward. Come forward quickly. Yes, please. We can move to the next lot. Yes. Next lot, please, quickly. Murarji Desai, Silver Oak, Yes, please. Please come forward. Yes, after this, all those schools who are left can come over, the students as well as the mentors. All those schools after this who are left can come forward, the students as well as the mentors. That would be the last camera shot for the day. So whoever is left, please come forward. There is a session scheduled after this. So please, whoever is left, come forward.

The students and the mentors. Quickly settle down please The last lot is here Thank you Ma ‘am you can Just settle down Just settle down this room Thank you ma ‘am Thank you Thank you Thank you ma ‘am Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you Thank you