Full Transcript

Dhruv Sharma: Hey there listeners, I don't know if you've seen it yet, but last evening we did a special covering Kunal Shah's ascension to the top job at WhatsApp. Now for today's stream, which is stream 104, Siv and I are chatting with Anshul Khandelwal, who's one of the co-founders of a company called Scent. I don't know if you guys have gotten your health checkup done for the year, but if you haven't, I think you want to tune into this conversation and get one of the best deals out there.

Anshul, welcome to the show.

Utsav Somani: Thanks, thanks a lot. Glad to be here. Thank you for joining us, Anshul.

So whenever somebody does healthcare or decides to tackle such a big problem, there is usually a personal story behind this. Do you guys have one? No, sure.

Absolutely.

Anshul Khandelwal, Co-founder of Cent: I think this is one area where everybody has a personal story. So I think my mom is a cancer survivor. She was lucky, she got detected at stage zero breast cancer.

Around four, five people in my family have passed away in the last four years of cancer, because of cancer. And more than half of them, two, three of them were unavoidable. And the rest of them, a couple of them were actually avoidable, and we could have saved their lives if you would have detected it in time.

And just recently, in the last three months or so, a couple of my cousins over 30 year olds have been detected with cancer. So I think it's, it's just all around us. And I think even with Shashank also, I think we tell this story often, very close friends and family have passed away, despite having done regimentally annual health checkups, despite that.

And when you figure out that it could have been avoided, it's really, it's not a great feeling. I think you live with that regret all your life. If only I could have done that.

That's the feeling. That's the mission. We don't want people to have that regret ever.

So yeah, that's the reason we just thought that the current healthcare system is not able to solve this problem. So what can we do about it? What's the best that we can do and bring it to India? What is SENT right now? Good question. SENT is a preventive health intelligence platform, which basically prevents and detects life-shortening and life-threatening illnesses, primarily across cancer, neuro, metabolic and cardiac conditions, which we call as CCNM.

These conditions are responsible for more than 70 to 80% deaths in humans all across the world. So all deaths, 80% of them are caused either by cancer, cardiac, neuro or metabolic. The way we do it is through a multimodal diagnostic platform, which basically combines very advanced imaging, pathology and genetics.

So all the data put together, which is then synthesized and orchestrated and fused with AI, along with a team of very highly educated and experts in radiology and medicine. In imaging, the biggest thing that we do is a full body MRI, which is very rare. MRI has always been used for curative purposes.

We are probably amongst a handful of companies in the world who use MRI for preventive screening. Full body MRI basically gets about a billion parameters in your body in one scan and could collect anywhere between eight to 10 GBs of data per individual. And that amount of data from head to toe basically is the gold standard in figuring out if you have anything in your body in simple ways which can kill you in the future.

And it's not just about life-threatening diseases, but predicting diseases which could happen in the next couple of years. And if you do it longitudinally, then you actually get to see how each part of your body is moving over a period of time and can visualize. For example, spine.

Most of us have spine issues. 80% of the people our age have spine issues. If you do it regularly, you can see the spine actually moving and how is it moving.

So it's not just about life-threatening conditions, but 300 plus conditions is what we test through the scan. And just to conclude, the product is a singular product. It's called the SEND scan, which is a two and a half hour process.

And you come into one of our centers, get in, get out. You would be the only one in the center, unlike any other diagnostic or hospital setup. And yeah, and then you get your reports in about 48 hours.

And then you do a 40 minute consultation with the MD doctor, who then walks you through every data point in that report.

Dhruv Sharma: That's fantastic. And give us a sense of scale, Anshul. How many centers do you have?

How many scans have you done? How many lives have you impacted already?

Anshul Khandelwal, Co-founder of Cent: Yeah, so we've been around for about six, seven months now, operationally. We've done more than 5,000 scans till date. We have three centers in India, one in Bangalore, Bombay, and Delhi.

And we would probably be present in about six cities in the next few months, and about 15 to 20 centers by the end of this year, which would be India's largest preventive health network or infrastructure for India. Of these 5,000 people, the most important part is 30% of the people we have found with a clinically meaningful finding, which essentially means the breakup of that is 25% of these people have found with something major in their body. 5% have been found with something critical in their body.

Critical essentially means if you don't take care of it right now, you could possibly die. Or you could look at it as if it is critical, you have to go to a OT right now, or you have to immediately consider going to the hospital and talking to someone. The rest 25% people is major, which is if you don't take care of it in the next six to nine months or one year, it could turn into something critical.

And the conditions to which we have found are brain aneurysms in normal 30-year-olds, 40-year- olds, extreme heart blockages, cancers. More than 2% of them are cancers all across the body. Significant spine issues, very, very high.

Visceral fat, extremely depleted metabolic function. So in fact, a 30-year-old walked into our center, he was a runner on a cycle very recently and walked out knowing that he has a cancer in his kidney. And he just came literally from a run.

Dhruv Sharma: This is eye-opening, Anshul. So the 5% with some sort of morbidity, did they walk through your doors because they sensed something was wrong? Or did they just...

Anshul Khandelwal, Co-founder of Cent: Sure. So that's the beauty, right? All of them, most of them, or when I say sorry, not almost, pretty much 90% of them were asymptomatic.

They had absolutely no clue of what is going on in their body. So I think the body is a great keeper of secrets, right? And that's the symptoms only start showing when it is practically really late.

And I think that's delays are only thing to solve and the only enemy also. And hence, most of the people who come to us are pretty much extremely healthy individuals. They are not symptomatic at all of anything critical. And the ones who find out something, for them, it is life-saving. So for the 5,000 people that we have done and the 5% critical cases, I would like to say that in some way, we have contributed to those 250 people in actually saving their lives.

Utsav Somani: And in India, screening culture doesn't exist. For most people, getting an annual checkup means that maybe running on a treadmill for like two minutes and taking a blood test, like all the simple stuff which all the hospitals sell and try to get you out of the hospital within a couple of hours and then give you a standardized report. So there's no personal care attached to this whole thing.

Is it because we are not educated enough about this whole setup or setups like yours haven't existed before? Why is India not believing in screening and preventive care?

Anshul Khandelwal, Co-founder of Cent: So two things I'll talk about. One is, of course, the system and second is the consumer psychology. Let me talk about the latter.

How much does the test cost, by the way? It's 27,500. It's an annual scan.

You do it once a year and then you repeat it annually as recommended. And then if you have to follow up, if we find something in you, we recommend you to follow up, then it could be six months also. So I think to answer your question, the second part, consumer psychology, I think that is not just Indian.

It is largely global, which is the optimism bias that if I'm feeling fine, I'm fine. And that's something which drives everything. But the bigger insight in all of this, which we have found, especially in India is I'm too scared to face the uncomfortable truth.

But at the same time, I'm also scared to find something late. So this creates this tension in the mind and the procrastination that if I tell you right now that this is what we do, you will say, yeah, I think we should do it. You might be thinking about it that I should do this scan.

But in all likelihood, you will not do it. You will procrastinate. So I think a lot of people are just and it's a very human thing to have there.

And that's also because of the memories or the experience of the semiotics of the Indian healthcare industry, where the moment you hear healthcare or a hospital, you automatically have imagined a place where sick people come, where you're sitting with 30 people in robes, in blue robes. And it's a place which scares you. And then you have to wait for three hours.

And then you have to go from one door to another. And it all plays out in your head. And that's why we have very, very negative sort of associations with healthcare, first of all. On the first part, I think the system itself was built to treat the sick. It was never built to look inside a healthy person. There is no reason to look inside a healthy person.

And which is why the entire system was built like that. So Indian medicine is great at obviously treating things, but we are extremely poor at anticipating something. And that's also, I would not blame, I think it's a function of we are a young country.

Till now we were dying of basic things. And now we have come so far as a country. And hence we've built a great health infrastructure to treat the sick.

But now is the time to build a completely new infrastructure so that people don't fall sick. And I think that infrastructure cannot be retrofitted. It can't be retrofitted in existing healthcare system.

It requires a new infrastructure layer to be built from ground up, which is what we are trying to do. And I think these are the reasons. One, the system didn't allow for this.

And second, the psychology. However, now there is a significant increase in one health awareness, of course. And it is actually the underestimate.

People like us probably think of it as Bangalore or Gurgaon, Tech Bro kind of ecosystem who are into health and biohacking and high rocks and all of those things. But as we have seen, the level of health awareness deeper into India is actually significantly improving. Our core consumer, you know, when we started, a lot of people told us that your consumers are the biohackers.

Versus the person who actually comes is a 45 year old guy who is a middle class, upper middle class family guy who is basically come to terms with mortality. As simple as that. Who has come to terms with the idea that my parents are getting old.

I have to take care of my family. I am 45 and hence I am now responsible and intelligent. And these are the people who are coming to us, not the people who track every calorie, who track REM in their sleep.

Those are not the people. They're of course there, but 90% of the people is average Joe, average Indian. And that's what I think is surprising, which is why the demand is so high.

We are actually constrained of capacity right now because there is so much load in India of existing diseases and awareness is also increasing. Just to give one more data point, I think we've gone through a significant, what you call as a epidemiological crisis, sort of transition in India, where two thirds of diseases are now non-communicable diseases versus 1990s where it was one third, right? And one fourth of them are in people in less than 70 years of age.

So you are essentially losing a lot of productive years of your life because of non-communicable diseases. And these are across cancer, cardiac, neuro and metabolic. And unfortunately, 70% of the cancers are detected after stage three, which effectively means cost is 10X and your prognosis is really bad.

Heart attacks, I don't need to repeat. Everyone knows about it. Everyone has read and heard, but men less than 50 years of age is 50% of the heart attacks.

So I think it's time that we at least do something about it. And the best we can do is look inside your body and it is giving you the signals and data and you do that every year. That's the infrastructure we are building.

Dhruv Sharma: Anshul, your earlier remarks, you mentioned three things. You said you're building a multimodal system and you mentioned three things, imaging, pathology and genetics. Can you talk a little more about each of those three things?

Anshul Khandelwal, Co-founder of Cent: Sure. So I think the first is imaging, which is essentially what is going on in your body from, of course, a deeper standpoint where we do five things. First thing is a full body MRI.

A full body MRI is necessary and important because MRI looks into soft tissues of your body, which no other imaging modality can. MRI is the gold standard of imaging. And the reason why MRI was never done was because it was extremely expensive.

It took a lot of time. If I were to do a full body MRI scan of you, it will take three hours. The reason why we can do it now is because of AI, I can now reduce the time to do a full body MRI to 30 minutes and that improves the utilization of my machine dramatically.

And hence, I can bring the cost down of MRI. And people always believed in it, but because of these reasons, never did it. So the first thing is full body MRI.

Then you do a cardiac CT, which essentially... So the point here is, and you would have... Sorry, I'm just digressing a little bit, but you would have heard about mid journey and all the things that mid journey has been doing.

I think the idea is that health is body is one thing where there is a reason why all these modalities exist, because no single modality can look into every part of your body. There are physics limitations to do that. Ultrasound, for example, cannot look into your bones.

Wherever there is bone or air pockets, ultrasound will not work. And it's physics. Nobody can actually change physics.

It's first principle. And hence, there has to be a multimodal approach to diagnostics, which is why we do it in the first place. We do a first full body MRI. We do a cardiac CT, which is a detailed heart imaging, extremely low dose in radiation, which is under all thresholds of recommended thresholds. Then we do ECG and echo. So ECG, echo and cardiac CT combined gives you a very detailed picture of your heart along with the blood parameters.

Then for women, we do a mammography. And for everyone, we do DEXA, which is India, especially it is extremely important because Indians have very high visceral fat and very poor body composition. So this is imaging.

So we get all of this imaging data. Then we do a 120 plus advanced blood biomarkers, not the natural ones, not the obvious ones. The other ones which are now much more advanced to even predict your risk of heart diseases, etc.

So we do a very advanced panel. And then we do genetics. Now, the correlation between these three, let's say if you have a gene in your body where you're more likely to have breast cancer.

If I find that out, then when I'm doing your MRI, I will possibly focus on that organ adaptively much more. So instead of doing it in silos blindly, you take all of this data and in real time can actually do adaptive protocols in imaging itself. Once all of this data comes is when, of course, we synthesize this with AI and it orchestrates and correlates everything and helps our radiologist to report faster and more accurately, which reduces the time of his reporting and improves his accuracy.

And then, of course, we publish an organ level risk report. So this report is basically a 70 page report, which is of your 10 to 15 organs, 10 organs in your body, every organ detailed out. Yeah, sorry if it is a long answer.

Utsav Somani: No, no, I think super informative. And I think interesting that you mentioned Midjourney's new venture into healthcare. Like, I mean, they're going from image gen to literally this.

And I read about Daniel, Spotify Daniel's Neko Health as well. And just closer to home, there's Fuji, I think, which is doing Nuda Health as well. So tell us a little bit about how you stack up compared to these players.

Like, what are they doing differently? How are you thinking about your company?

Anshul Khandelwal, Co-founder of Cent: Sure, see, for us, I think the north star metric is accuracy. There are other metrics to accuracy, which is false negatives and false positives, which is the biggest risk of doing screening at a population scale. You will end up having more false positives and false negatives is, of course, then you're a bad product if you have false negative. Everything we do optimizes for this metric, which is how accurate I am, which leads to more quality adjusted life years to you. If I were to compare the protocol to anyone else, I think a lot of them are very specific to certain parts of the body or certain number of conditions. Like I said, ultrasound, everyone is talking about it's a great product, but there is a lot of skepticism around it also because the technology cannot, one technology cannot really understand the whole body is the current understanding.

And which is why I would say it's a really cool product. It will help solve for 50% of maybe the conditions up to your torso from overall from here on. It can't look into the brain.

It's very, very hard for ultrasound to look into the brain. And if you have brain aneurysms, because it's the skull is born. So hence it's difficult.

So I would say everybody is trying to solve a sliver of the problem. And of course, the entire ecosystem put together, there will be a lot of learnings from here on. So I don't know how to compare, but we are very focused on the science, science of accuracy.

And that's one metric we will continuously keep optimizing for. I also feel one important point is, is nobody has ever collected data for asymptomatic individuals. All the data that exists is for symptomatic individuals in the healthcare industry or everywhere.

There is no AI only knows sick people. It doesn't know healthy people. So it doesn't know what is going on in a healthy people, a healthy person at a 30 years of age and at 35 and at 38.

So what we are doing is essentially building this very unique Indian dataset of asymptomatic individuals. And once we have this data, which is why we want to open up these many centers so that we can give access to people, collect as much data as we can. And then the magic happens.

Right now, of course, if I have scanned 5,000 people, it's not enough. But once I have 100,000 people, for example, I will have the largest data set of healthy Indian bodies. And that's something which nobody has.

So I think the objective of screening programs is this. And I think everybody is doing their bit and feels like it should become a norm in the near future, where just like your life insurance and health insurance every year, it will become a norm and a ritual to do it every year. And it will be foolish if you don't do it.

Dhruv Sharma: So can I maybe pull you back from the science a little bit and ask about maybe marketing and design problems. So for instance, in the course of the conversation, you said that a lot of people have come to you, these 5,000 people, they've come to terms with their mortality. I think on a base of 100, maybe there's five such people. 95 people would want to walk through your doors and walk out of your doors with a clean bill of health, right? Maybe I'm one such person.

Anshul Khandelwal, Co-founder of Cent: Absolutely.

Dhruv Sharma: So how do you market something which could in the short term deliver bad news without appealing to fear in people, but rather appealing to hope, right? That there's a cure for most of these things. That's one.

And design because your spaces also have to reflect that, right? I don't know, more natural lights and plans for life, signs of life, et cetera. So how are you designing your space as well?

Talk about both of those things.

Anshul Khandelwal, Co-founder of Cent: I wish I could give you a tour of the space right now.

Dhruv Sharma: We'll do one later.

Anshul Khandelwal, Co-founder of Cent: It's absolutely bang on. That's the design principle. If you walk into the space, I don't know, I don't like to use the word, but you would feel like a spa.

It's an extremely warm space with a lot of greenery, natural light. Like you said, the idea is to keep you to two things. No anxiety because anxiety leads to stress.

Stress is not good for your body. So the place is extremely welcoming and warm. There's no other way for me to define the aesthetics.

I think once you look at it, but I would like to say it's the biggest compliments that we get is on the space itself. People feel really at ease and spend three hours here and have that.

Utsav Somani: Do you think yourself of Scent Healthcare as a premium consumer brand or as a healthcare company?

Anshul Khandelwal, Co-founder of Cent: I think we're a premium. I feel is extremely, I dislike that word. I don't know what does premium essentially mean, but we are a consumer healthcare company and we are direct to consumer.

The one thing I forgot to tell is in our business model, it's one of the first ones where there is absolutely no involvement of hospitals, insurance companies, diagnostic labs, referrals. All the money that we make, all the relationships that we have is direct to consumer and which is even harder. I could have had the easy job of getting people, that's how the Indian healthcare system works or diagnostics industries work.

But we took obviously the hard part because that is the ethos and there is a conflict of interest if I do otherwise. And trust is the biggest thing in healthcare. So we directly communicate and it's hard to, of course, for all the reasons we discussed to persuade a large mass of Indian population to take care of themselves and their bodies.

So yeah, I would like to call ourselves as a consumer, direct to consumer Indian healthcare.

Dhruv Sharma: And like Dhruv, when you asked space- No, I love the analogy, like you comparing it to a spa, right? Like hell, if I'm walking in to get such an extensive scan, I even want a peanut butter smoothie or something by the end of it, right?

Utsav Somani: Yeah, yeah. Should we spend the next minute, the last minute that we have? Anshul, do you wanna just maybe show us the lobby?

If not, any particular room? I'll just take my laptop and show you.

Anshul Khandelwal, Co-founder of Cent: Okay, so how do, okay. Oh, wow. This is the Bangalore- Sorry, sorry.

I don't know how to reverse the camera. One second. Guys, can you help?

Utsav Somani: It's okay. I mean, you can just use the- Hi, everyone. Oh, nice.

This looks very good. This is the Bangalore center? Yeah, yeah.

It's the Bangalore center. And how many square foots?

Anshul Khandelwal, Co-founder of Cent: This one is 7,000 square feet.

Utsav Somani: And that's, I mean, roughly the space that you need to set up a single purpose center.

Anshul Khandelwal, Co-founder of Cent: No, no, no, no. This one is, of course, our first one. We have two floors here, 7,000 square feet, but we can do it in about 5,000 square feet.

Utsav Somani: All right, Anshul. This was really, really informative for both of us. Thank you so much.

And where can our listeners book a test for themselves?

Anshul Khandelwal, Co-founder of Cent: Yeah, just log on to Send.Health. That's it. All right.

Utsav Somani: We'll keep it simple. Thank you so much for joining us.

Anshul Khandelwal, Co-founder of Cent: Thank you, sir. Thank you.

Utsav Somani: All right, listeners. We're moving on to our next guest. We've got Sarthak from Miticata.

Sarthak, welcome to the show. Thank you. Thank you so much, sir.

Thanks, Rov. You know, thanks for calling me over here. No, no, congrats on the new milestones and announcing the round.

So thank you for giving us the time and giving us a chance to break down the headlines with you. I mean, let's introduce the business first. What does Miticata do?

Sarthak Dubey, Co-founder & COO of Mitigata: So Miticata is essentially a full stack cyber resilience company. We manage the entire cyber risk for the business, whether it is cyber risk prevention, cyber risk mitigation, or cyber risk transfer. So the prevention and mitigation piece comes with cybersecurity products and the compliance products that we have built in house.

And the risk transfer bit happens via cyber insurance for which we work with all the insurance companies or the reinsurance companies across the world. So essentially we cover the entire element of the risk and entire loop of the risk where the business does not need to work with 400, 500 different vendors to manage their different cybersecurity needs separately. They just have a single point of contact.

They have a single point of connection to manage anything and everything related to cyber. And that is what Miticata does.

Utsav Somani: And what's the scale of the business right now? And when we first spoke, and of course I'm an investor in the company, so thank you for taking me along for the ride. You started off with insurance, but you eventually, I mean, as you came to this series B, 15 million round led by some top names, how's the vision evolved or developed?

Sarthak Dubey, Co-founder & COO of Mitigata: Yeah, absolutely. But before I answer that, I mean, for all listeners, I think to Dhruv, everyone, Utsav was the decision, like the meeting that we had with Utsav before Utsav decided to invest in us was probably a five minute or seven minute discussion. And then he just responded saying that, oh, we are just doing it.

This was the quickest investment that I had ever received. So from a business perspective, while of course we started with cyber insurance, currently it's cyber insurance, security, compliance and AI SOC which contributes to equal part of a business. So close to somewhere around 20-25% each of the categories contribute to our business.

Dhruv Sharma: And Sarthak, of course, we have to ask you the Mythos question. What the whole charade was all about.

Sarthak Dubey, Co-founder & COO of Mitigata: So essentially Anthropic created a very advanced model, Mythos. The capability of that Mythos was in some sense much more than what a human security researcher could have identified or done. It was actually able to identify certain vulnerabilities in the system that none of the human has been able to do before.

Mythos would have taken that couple of steps at least 15 to 20 steps ahead. Until and unless we are building something in-house in India, there is a sovereign AI model, there is always going to be sovereignty risk.

Utsav Somani: I think at this point, every country should just rename their sovereign AI project to the Manhattan Project for AI.

Sarthak Dubey, Co-founder & COO of Mitigata: Yeah, this is a new nuclear arms race. For sure.

Utsav Somani: When should businesses come to you and how do they realize they have a vulnerability?

Sarthak Dubey, Co-founder & COO of Mitigata: There are two ways to look at it. There are one set of companies which are in the regulated space — the regulations itself make you do it at the get-go. For a lot of businesses in the regulated industries, even if you are a small startup, we have solutions for them.

As soon as the company starts growing and starts making any sort of money or having any kind of data, any kind of users, they are at risk.s would pinpoint a company and then run certain attacks on them. But what is happening right now is because of AI coming and because of the AI agents coming and because of the speed of which the attackers attack companies has become very fast.

From a business perspective, we cover whether it is a very small bootstrapped company or multinational companies. We have solutions for each kind of client. It can start with as low as Rs 50,000.

Dhruv Sharma: Which layer in the technology stack is most exposed to threats?

Sarthak Dubey, Co-founder & COO of Mitigata: The majority of the attacks originate from the human layer — basically your vendors and your employees. Those 40-45% of attacks are originated from employees themselves, not even a third-party vendor. Phishing attacks have now become very sophisticated with AI making emails look exactly like originals.

One of our clients — an engineer received an email from Atlassian for a party with free drinks. He ended up logging in with Google and the entire company ended up spending around 40 lakhs for their entire security mess that followed. Something as simple as phishing is still the classic and most effective way.

Utsav Somani: Final closing question, Sarthak, what's next for Mitigata?

Sarthak Dubey, Co-founder & COO of Mitigata: International expansion — Middle East and Southeast Asia are the two geographies we're targeting. These markets are very similar to how an Indian company thinks about cybersecurity — very young in terms of maturity. Apart from that, we are increasing our capability and working very closely with our R&D teams on AISOC. We already are by far one of the most efficient in the country to be running AISOC operations.

Utsav Somani: Awesome. Wishing you and Mohith all the best. Thank you so much for coming on our show.

Sarthak Dubey, Co-founder & COO of Mitigata: Thank you. Thanks.

Utsav Somani: Alrighty, that's it from the guest segment today. There's a quick readout about the InfoEdge shareholder letter. I think fascinating that they started in the world of AI and deep tech fairly early around 2020. Invest early before consensus and stay invested longer than most others. They've invested over 1000 crores, 1003 to be precise, in 54 AI companies and deep tech startups since 2020. AI companies account for approximately 614 crores deployed with 31% IRR. With that, we'll sign off. Go read the letter. It's a fascinating read.

Dhruv Sharma: There is something about InfoEdge's model that allows them to do this. They're not a fund. They're not bound by the term or life of their fund. They can pretty much hold an asset in perpetuity. The SpaceX IPO just happened — it was 24 years in the making. That's two and a half fund cycles. Not everyone's cup of tea to stay with the position that long.

Utsav Somani: Awesome, awesome. We'll see you on Friday at 4 p.m. Thank you so much for tuning in. Bye-bye, folks.

Utsav Somani: And what's the scale of the business right now?

Sarthak Dubey, Co-founder & COO of Mitigata: So Mitigata is essentially a full stack cyber resilience company. We manage the entire cyber risk for the business, whether it is cyber risk prevention, cyber risk mitigation, or cyber risk transfer. The prevention and mitigation piece comes with cybersecurity products and the compliance products built in house. And the risk transfer bit happens via cyber insurance for which we work with all the insurance companies or the reinsurance companies across the world. So essentially we cover the entire loop of the risk where the business does not need to work with 400, 500 different vendors to manage their different cybersecurity needs separately. They just have a single point of contact to manage anything and everything related to cyber.

Before I answer that, for all listeners — the meeting we had with Utsav before he decided to invest in us was probably a five to seven minute discussion. Then he just responded saying that we are just doing it. This was the quickest investment I had ever received.

The idea was always the same: cyber insurance as a standalone product cannot exist because the amount of risk is very uncertain. Very limited data points in India because nobody goes around telling that they have been attacked. So currently India is in a space where probably health insurance was 25-30 years ago. Because of this, it became very critical for us to combine cyber security and insurance together.

At this point the business is very equally divided — cyber insurance, security, compliance and AI SOC each contribute close to 20-25% of the business.

We built something called Gordon AI and Gordon Console that gets telemetry from across the organization — devices, tools, third-party vendors. A CISO who starts his day can simply ask: who are my top 10 riskiest employees based on EDR data from Palo Alto, SIEM data from Splunk, and their last phishing score? And automatically run training or warnings. Before this, a CISO would spend an entire boardroom discussion just to understand the problem.

Dhruv Sharma: What is telemetry data in this context, and how does it help quantify cybersecurity risk?

Sarthak Dubey, Co-founder & COO of Mitigata: Telemetry data are logs that identify, in case something goes wrong, what exactly were the initial evidences that led to that vulnerability getting exposed. This is very similar to your body — your body starts showing signs when something is wrong.

An employee using the company's email ID to log into Swiggy — this itself is a data point that nobody knew about, which is actually increasing the risk for the business. Before, the data points existed but nobody did anything about it.

Dhruv Sharma: Tell us about the whole Claude Opus situation — just assume we know nothing.

Sarthak Dubey, Co-founder & COO of Mitigata: With the power that Anthropic created with Claude Opus, the capability was in some sense much more than what a human security researcher could have done. It was actually able to identify certain vulnerabilities that no human had identified before. If a black hat or threat actor gets access to something like this, it could result in catastrophic failure of the entire security infrastructure. Threat actors are always a couple of steps ahead of security researchers — Claude Opus would have taken that at least 15 to 20 steps ahead.

But where does it stop? At what level of infrastructure — whether security, AI, or defense — is the power given to Washington DC to just shut off a button? Until and unless we are building something in-house in India — a sovereign AI model — we need to build our own sovereign infrastructure. Sarvam is doing a really good job but all together, we need more.

Utsav Somani: Every country should just rename their sovereign AI project to the Manhattan Project for AI. It seems like every country is going to race towards having their own nuclear arms in the world of AI.

Sarthak Dubey, Co-founder & COO of Mitigata: This is a new nuclear arms race. For sure.

Utsav Somani: When does an Indian business realize they have a vulnerability? What kind of businesses do you work with?

Sarthak Dubey, Co-founder & COO of Mitigata: Two ways to look at it. Companies in the regulated space — BFSI etc. — are forced by regulations to have certain guardrails from the start. For companies outside regulated industries: as soon as a company starts making money, having any data or users, they are at risk.

Currently, AI has made the speed at which attackers move very fast. Before, attackers would pinpoint a company and run attacks. Now they cover multiple companies in a single attack. We work with clients from bootstrapped three-person companies to multinationals with hundreds of thousands of crores of revenue.

For small companies, it can start with as low as Rs 50,000. But that changes everything — these are preventive measures. The reactive measures — forensics, incident response, ransomware recovery — cost much more.

Dhruv Sharma: In the entire technology stack, from human layer to application layer to kernel, which layer is most exposed?

Sarthak Dubey, Co-founder & COO of Mitigata: 100% correct that it gets progressively harder as you go deeper. The majority of attacks originate from the first line — your vendors, your employees, the human layer. 40-45% of attacks originate from employees themselves, not even a third-party vendor.

Basic phishing attacks, which have now become very sophisticated with AI-generated spoofed emails. One of our clients — an engineer received an email from what looked like Atlassian for an event with free drinks. He logged in via his Google credentials and the entire company ended up spending 40 lakhs. Something as simple as phishing is still the most effective way to get into somebody's system.

Utsav Somani: What's next for Mitigata?

Sarthak Dubey, Co-founder & COO of Mitigata: International expansion — specifically Middle East and Southeast Asia, which are very similar to how Indian companies think about cybersecurity: young in terms of security maturity. Apart from that, increasing our AISOC capability. We are by far one of the most efficient in the country running AISOC operations. Improving R&D, product, getting the right people, and expanding geographies.

Utsav Somani: Wishing you and Mohith all the best. Thank you so much.

Sarthak Dubey, Co-founder & COO of Mitigata: Thank you so much.

Utsav Somani: That's it from the guest segment today. There's a quick readout I want to do about the InfoEdge shareholder letter that just came out. Did you get a chance to read it?

Dhruv Sharma: I've seen some snippets, I haven't read in depth. I'd love to hear.

Utsav Somani: And you know Sanjeev personally as well. You hosted him for an AngelList India event.

Dhruv Sharma: That's right. Yes.

Utsav Somani: Some fascinating stats. They started investing in AI and deep tech fairly early around 2020 before the product market became active. There's a statement: invest early before consensus and stay invested longer than most others. He's shown this over Policy Bazaar and Zomato — he's still holding everything.

They're investing from the balance sheet, most of the capital as LPs in the funds they run. They've invested over 1,003 crores in 54 AI companies and deep tech startups since 2020. Portfolio includes LAT Aerospace, Temple, Gnani, Eplane, Go and Attentive AI.

AI companies account for approximately 614 crores deployed with 31% IRR. And he makes a statement: be skeptical of interim IRRs. Markups are only temporary until IPOs or exits happen. So it's DPI, DPI, DPI for everyone in the venture world these days.

Go read the letter. I think it's a fascinating read. We might be getting close to our own version of Warren Buffett letters soon.

Dhruv Sharma: Absolutely. There is something about InfoEdge's model that allows them to do this. They're not a fund. They're not bound by the term or life of their fund. They can hold an asset in perpetuity.

The SpaceX IPO just happened — it was 24 years in the making. That's two and a half fund cycles. Not everyone's cup of tea to stay that long. And as Gaurav Sheth from Pearside has noted, very few people know how to underwrite and price failure. Building in space is going to involve failures in development, at launch, in orbit, at reentry. Everyone is going to have to wrap their heads around a very different problem statement.

Utsav Somani: We'll see you on Friday at 4 PM. Thank you so much for tuning in. Bye bye, folks.