Are we good? Ready guys? Should we start?
You've done so many interviews now. I never felt so intimidated. There is so much information around the India Stack that is so relevant to a young person who wants to be an entrepreneur in India. Hopefully by the end of it, we can learn from you what should be built on top of the India Stack and what is the opportunity today from the lens of capitalism. Hi Anandan, thank you for doing this.
My pleasure, Nikhil. I don't think many people know, but I end up meeting Nandan most often in his house or when he's traveling to one of 100 different countries that he goes to every year. That's right. So where are you going next, Nandan, in your travels? Actually, this month is a slow month in the sense I'm going to be right here in Bangalore and in India.
But I'm going next month. I'm going to Basel, I'm going to London. All around this technology broadly that you're involved with? Yeah, I basically have two hats. I have a public hat, which is really the bulk of what I do, which is enabling digital public infrastructure.
You know, Aadhaar, UPI, that kind of stuff. And now we are trying to make that go global. So part of this is the globalization of this DPI concept.
And then I have a private sector hat as the chairman of Infosys and a few investments. So some of my travel is that. Or if I go on a vacation, I just work for a couple of days.
I watched a bunch of your interviews. Why would you do that? Over the last 48 hours.
You seem to be a sucker for punishment. I feel like I know Nandan, the person in private. I have never heard you speak at length publicly in that way.
I've heard you speak at your house when you host dinners and all of that. So it was actually intriguing and there is so much information around the India stack that is so relevant to a young person who wants to be an entrepreneur in India. So we thought we'll focus today on that. Sure. So I'm going to play the part of a 20-year-old who knows nothing about the India stack.
I'll start the journey from the very beginning and hopefully by the end of it, We can learn from you what should be built on top of the India stack. And what is the opportunity today from the lens of capitalism? What can me, a 20-year-old, start to make money out of it on top of the India stack in any way?
If we were to begin with emphasis, is that where you would say the journey began in that manner where you saw scale incrementally? Yeah, I think Infosys was a great learning journey in scale. You know, Infosys was founded in 1981 under the leadership of Mr. Narayan Murthy and I was one of the co-founders. You're my favorite co-founder. I know a bunch of them.
Okay, I mean, well, anyway. So I was part of that journey for a long time. And I learned...
a little bit about scale. That's enterprise scale and population scale is a completely different ballgame. So enterprise scale, how to think, how to sort of look at what the future can be and sort of drive something towards that future.
And then that's how I, you know, I was still, 2009 I was very much in Infosys. And then I got this offer to join the government and lead the Aadhaar project. Infosys was the start of the IT boom in India, right? And also Bangalore became what it is today.
Yeah, I think you can think of Infosys as the original startup. Right. You know, we moved to Bangalore in 1983. From? From Bombay.
Initially we were in Bombay, Pune, in that belt. And we've been there ever since. Bangalore has had a series of technology curves, hardware, software, internet. You know, so many things have happened.
But yeah, we were part of that wave and I think we played a big role in establishing Bangalore as a tech city. And also, I think even today, I meet so many people who are entrepreneurs who said they were at Infosys. Right.
So I think it also spawned so many other people. What worked for Infosys in a world where the incumbent large business houses were typically family businesses or I wouldn't like to say crony capitalism, but some kind of a political element involved in a business at scale? Right.
What set Infosys apart at that point of time? I think the group of founders were very synergistic. We all had different strengths. Muthi was our leader. We had common values.
We wanted to build for the long term. We were marathon runners. We're not sprinters.
We built for the long term. We had a vision of creating a globally respected company of having very high standards of... corporate governance. So the whole set of values, Infosys values, which are deep in everyone.
And that I think sort of gave the fuel to make it happen. You have to have a vision, you have to have passion about something, you know, we had all that. And you moved from Infosys, you retired first and then Aadhaar came your way?
No, what happened was in 2009, in 2008, I wrote a book called Imagining India. So sort of projecting how India could be in the framework of ideas. Why ideas matter and stuff like that. And among the ideas in that book was digital ID and so on. So at the same time, the government was planning a digital ID program.
And they had got cabinet approval for that project. And I had a chance to take up that role. And then my Infosys co-founders were very kind and said, yeah, you can go and do some national service.
So that's how I ended up in Delhi in July of 2009 to build India's digital program, ID program. And before Infosys Nandan, you were in Bombay before that? Yeah, I was born in Bangalore.
In that sense, I'm an original Bangalorean. I was born in Maniwala's hospital in Bangalore. And my father worked in a...
textile mill in Maghdi Road. Right. In Maghdi Road.
Industrial area. Yeah. And then the, you know, the mill sector got into difficulties.
So he had to leave and he thought his job would be very, wouldn't be in a fixed place. So I got sent to Dharawal to be in my uncle's house. My uncle was very kind.
He took me in and I stayed there for five years and then went to school in Dharawal. College also. And then I went to IIT Bombay. And then from IIT Bombay, I joined a company which then, we were all working together and then Infosys was set up. And you met Rohini in college?
I remember you guys mentioned this story. I met Rohini because she was in Elphinstone College. Right.
I was in IIT. Yeah. And she described you as this badass person who did something when she narrated the story. No, no. What happened was, we had a quiz competition.
So I was representing IIT in the quiz and the quiz was in Elphinstone College and obviously we won the quiz. Why obviously you won? I mean we had that belief that we could be good at this stuff. So that's how I met her.
Way back in 77. She's the more outspoken of the two of you, right? She's more spontaneous. She's more, I would say more...
emotional or mercurial or whatever. I'm more sedate. I'm the boring type. And she says what she's thinking right away, impulsively. She's obsessive about climate in particular, right?
Environment is a big theme for her. Part of her philanthropy. She's the biggest probably environment philanthropist in India now. Do you agree and resonate with that belief?
Oh yeah, absolutely. I mean, I'm not that involved. But I completely agree with what she's doing. Okay, so if we were to start the IndiaStack journey, I'm going to write it down because I'm learning with everybody who's going to watch this. So, Aadhar, first thing first.
So, the government asked you to build a digital ID. No, the government asked me to build a digital ID. a unique ID. A unique ID. There was no digital in it. I made it digital.
Right. Because I said in 2009, if you're going to build an ID, it has to be a digital ID. Okay.
It has to be online. It has to be verifiable online with real time and all that stuff. So those were the value add that my team and I did was how do we...
Execute this using a digital first way. So in 2009, what was the benchmark you set for yourself while starting the Aadhaar project? Who were you looking at?
Who had done well with the unique idea? Nobody had done it. So it was a brand new idea. Is social security in America a similar concept?
Social security in America came in 1936. Right. Before technology. And in fact, the first use of mainframes was So mainframes were used because there was SSN. So SSN was a number, but it was a paper-based number.
Just tell us full forms of all of this. Yeah, so SSN is Social Security number, which the US did because what happened then was that Franklin Roosevelt, during the 30s after the whole economic crisis in the US, they came out with a scheme where people could get benefits in their old age if they worked. And then they had to keep track of a person throughout the age, right?
Because they would contribute while they were working and they would get their benefits when they retired. So that's how they needed to have a number. That's how the SSN came about.
Is that the main purpose with which governments'unique IDs, benefit transfer? No, benefits transfer is a big part of that. In our case, in the Indian case, there were two drivers.
One was efficient benefits transfer. Because since 2000, India has built more and more of a welfare programs, which meant you send money to people. And they had to have a solid, robust way of doing it.
So one part of this was to make benefit transfer more efficient. And therefore make... government targeting more efficient for benefits. The second thing was an equally important driver, which was that most people didn't have IDs. Because in many societies, like in the West, 98% of births are in a hospital.
And they get a birth certificate when they are born. So that becomes the root document from which you can then get other documents. In India, births were not happening in a hospital, they were happening at home, they were happening in a village. So half the babies born in some states didn't have a birth certificate.
so which was fine as long as you lived in a particular area but if you started moving around you know migrating going to the big city to enter a car bus or to get a job or if the policeman stopped you on the road you had to show some id and most people didn't have an id so that was a very fundamental impediment for their progress so inclusion of people into the formal system with an id and making benefits transfer more efficient whether government drivers of this initiative. Right. And did the market play a large role in this process?
No, we were looking at how do you create digital infrastructure on which market or private innovation could be done. Right. So, and we call this as public rails and private innovation.
This is broadly DPI. DPI. So the ideas evolved, but it was actually constructed on concepts like, I'm not on the internet.
Look at the internet. Internet was funded by the US government. But on top of that, you had all these companies that came up.
Similarly, GPS was again created by the US Department of Defense. But on top of that, we had Uber and all these things, Ola and so on. Maps. So the notion that if you create a digital public infrastructure at population scale and open it up with application programming interfaces so anybody can build applications, then it will unleash innovation.
So that was the philosophy of the design of Aadhaar and subsequently UPI as API-led architecture which everybody could use and then whether it's a startup or a large company they could then build useful applications on top of it. Right. And when you started on the Aadhaar project, this was 2009? Yeah.
How long did it take? Well, I was in the government for about four and a half years. And I had made a commitment that I would deliver 600 million people with Aadhaar by the time I quit. And I did that. So when I stepped down, I had 600 million people had an ID.
And then, of course, it continued. with the new government and now 1.3 billion people have the idea. So after Aadhaar came, E-K-Y-C-A and E sign on top of it?
Yeah, no. Aadhaar was an ID that just said, Nikhil is Nikhil. Right. And this happened between 2009 and 2013?
Yeah. And scaled up to 600 million. Okay. And with an online authentication. So you could go somewhere, put your finger and verify that Nikhil is Nikhil.
You say, my number is 123 and the system would confirm that it is that. So we call that an online authentication. Okay.
And then everyone's ID had the name and address and date of birth and all that. Mm-hmm. That was considered sufficient KYC to open a bank account, get a mobile connection or buy a...
What does Aadhaar have? Name, date of birth, mobile number, address? Sex, email ID and phone number.
It's quite simple but it was unique. A person could have only one number. The design was how to make it unique.
So then using the additional information in Aadhaar, you could do a KYC. And then the regulators agreed that this KYC was adequate to get a bank account or a mobile connection or buy a mutual fund or whatever. How did that work? How did you convince the regulator to accept Aadhaar as a... Yeah, so we had to first work with the government of India under the, you know, many acts.
Because at that time, you know, Prevention of Money Loaning Act and there was lots of things happening. So we had to get it approved as a KYC in general. Then we had to work with each regulator.
So we had to work with the RBI to get them to accept it for banks. We had to work with DOT and TRAI for mobile phone, IRDA for insurance, SEBI for capital markets. So we had to work with each regulator and we got all of them to accept it. And how much time did that take? That took maybe, it took several years, two, three years.
But the important thing was that the... Growth of that category happened at different times. For example, the use of Aadhaar KYC in banking happened after 2016. Or 2015 rather. 2015 when Prime Minister Modi wanted to do the Jandhan Yojana program, which was to give bank accounts to everybody.
And that was a big push from the government to issue bank accounts to everyone. And Aadhaar KYC accelerated that because you could do a KYC. immediately and get a bank account.
You mentioned somewhere that it brought down the time taken to give people bank accounts by from 47 years to nine years. Yeah. So on the normal GDP, if you had looked at it, it will take 46 years to get to the level of banking penetration that we did in nine years because you accelerated with digital technology.
EKYC was the reason for that? Huh? EKYC was the main reason for that? Yeah.
I mean, no, no. I mean, there was political push, government push, customer demand. Mm-hmm. Banks had to execute.
So a lot of things went into it. But the technological capability of doing a KYC online, on the fly, within two minutes, reduced your transaction time and cost. So it was an enabler.
Same thing happened with mobile connections. So we launched Aadhaar KYC in 2012-13. But it's only when Jio got launched in 2016 that they used Aadhaar KYC to enroll a million customers a day. Wow. Yeah, because they wanted to get to 100 million customers in six months.
Only way to do that was to use Aadhaar KYC. So they used it and the rest was history, so to speak. So they popularized the use of Aadhaar KYC in mobile industry. So Jandan did it in banking, Reliance Jio did in mobile, and of course, Zerodhaar did in capital markets. So you guys used it to get...
We are big beneficiaries of EKYC. And did eSign happen at similar time? Yeah, eSign happened in 2015 when Ramsevak Sharma, who was my DG when I was the chairman of UIDA, he became the IT secretary. So he launched eSign and DigiLocker and all these other capabilities. They came together, eSign and DigiLocker?
Yeah, they came. 2015, they were launched at different times. But I would say they launched at the same time.
So let's say DigiLocker is a... Wallet for government IDs? No.
It's a general purpose omnibus wallet for all documents. In a secure way. Which you can keep on your phone or on the cloud.
And you can keep your driver's license, your Aadhaar card, your PAN card. But it can be used by private guys too. How do you describe that?
General purpose? It's a general purpose document wallet. Secure document wallet which can be in the phone or on the cloud.
Which then... You can keep all kinds of documents. You can keep your high school graduation details. You can keep your marks card. It's not, it can be any document.
So now there are billions of documents in this thing, more than... And what did people use eSign for? eSign was to do digital signatures. So, you know, for example, if you buy a property and you have to sign there, you can do a digital signature or if you're signing a contract, you can use digital signatures. And you couldn't do this up until now because you didn't have a unique ID and you could not authorize the unique ID to sign, but now you could?
Yeah, yeah. A number of things, yeah, but effectively that. So eGAYC came in 2012-13, eSign in 2015, and DigiLocker in 2016? No, they both came in 2015. 2015. But the big year was 2016. What happened in 2016?
It was a remarkable year for our digital transformation. First on April 4th, 2016, India reached 1 billion Adhas. That was a landmark achievement.
April 11, 2016, India launched UPI. Completely independent coincidence. September 5th 2016, Reliance Jio was launched.
November 8th 2016, demonetization happened. And December 29th 2016, the prime minister launched the BEAM application. So in one year everything changed. Why do you add Reliance Jio to this list? Because it transformed India's mobile industry.
They launched very low cost, lowest data rates in the world. Voice was free. That created huge demand for smartphones. So smartphone prices fell.
You can see the data, they're very clear. In 2016 onwards, we saw the boom in data. Now before that, we had a voice economy.
Pre-Geo, we had what I call minute factories. Right. Very cheap voice calls. Right. And you know, but they didn't have data or their data was expensive.
What Geo did was completely collapse the price of data by creating a world-class 4G network and made voice free. So it changed the whole economics and demand perception of telecom in India. It's a huge achievement.
And what did Veeam do? Bheem essentially popularized online payments because Bheem was an app launched by NPCI, launched by the prime minister, built by NPCI. And for the first time... NPCI is? NPCI is the National Payment Corporation of India which runs India's payment systems.
It's an amazing organization. So NPCI launched Bheem and Bheem was the first exposure people got to a mobile payment app. And then of course, everybody came in, you had Google Pay, Phone Pay, Paytm, all that. And then of course today, UPI does 14.4 billion transactions in a month. Can you explain the difference between a Bheem kind of transaction and a UPI transaction?
A Bheem is a particular app. So it's a consumer app on my phone on which I can initiate a payment. UPI is Unified Payment Interface, which is the underlying protocol which makes the payment happen between two apps and two different bank accounts.
Can you break that down for me? What do you mean by protocol? I read a bunch of the stuff you've written as well where you use the word protocol a lot. What is a protocol?
Protocol is just a way of describing the packet of exchange of a transaction. So UPI protocol is for payments. So you're saying exchanging of information, let's say I have ABC I want to tell you, you want to tell me DEF, that exchange is called a protocol?
No, that exchange is done through a protocol, which is a way of describing. If you look at the internet, the internet has a protocol, TCP IP, which describes how data moves on the internet. So UPI protocol is a protocol that describes how payment values move in a payment network.
And that enables dissimilar organizations to talk to each other for that category of activity. What do you mean by dissimilar organization? Like two banks and two mobile apps.
For example, I am on Bheem. I'm in State Bank of India. You're using PhonePay or Google Pay and you are in HDFC Bank. I can pay you in real time. Why is that?
Because the protocol Everybody talks the same. It's like there's a language of that business, of payments. Would I be right in saying if UPI is the protocol that transfers value from one store to another? and its first representation is INR and the first store is a bank account, that same protocol can be used for other things? Yeah, it can be a multi-currency protocol.
It can be for wallets, bank accounts. Tomorrow it can be used for CBDC. So you can, once you have that basic infrastructure on that, you can add different types of accounts. So UPI will work as seamlessly for, let's say, remittances coming into India.
That requires a little more work. For example, India, the NPC of India under the ages of the RBI and Singapore under the ages of Mass, which is their central bank, they have built a connection between UPI and Singapore's fast pay or whatever it's called. That now I can transfer money from my bank account in a Singapore bank to somebody in India.
convert it from Singapore dollars to Indian rupees. Yeah, but you have to build those bridges. So it's a bit cumbersome because you have to go to each country and the two countries have to come out with the rules. So would you say that bridge is like a translation between two protocols? Yeah, but it's also about things like, you know, fraud, prevention of money laundering, capital account convertibility, a number of issues when you do a cross-country, right?
I mean, India, for example. Indians can't transfer more than $250,000 a year, LRS and so on. So whereas Singaporeans have capital account convertibility.
So how do you match the different rules of the two countries? A lot of the work goes in that. Right. So tell us more about UPI because it seems like it's this thing that changed the entire ecosystem. So you said it's a protocol, INR representation, store was one bank account to another bank account.
Why did it scale in the manner that it did? What did it provide that a traditional bank account to bank account transfer did not? Well, at that time, there was, before UPI, there was no way for two people to make a payment to each other. I mean, if I said to you, how do I send you money? There was no way to do it.
What if I go to my bank and do, they used to have the net transfer and all. Yeah, you can do net banking and all. You know, you go there, do something. Here I can just tap on my phone and send you money.
The convenience, real-time ease of use and convenience was a different league. Now, UPI was conceived in 2013. And in 2015, I became an advisor to NPCI on innovation and public policy. And NPCI built it. It took three years to build it and launched it on May 11, 2016. And it was designed by NPCI, Dilepasbe and others.
and Pramod on the architecture side, Dr. Pramod Verma was also the Aadhar architect. So all these people got together and designed this amazing, very high volume, very low cost, highly scalable payment infrastructure. So it made sending money as easy as sending an email or sending a WhatsApp message. Is that because now each person had a unique ID per se? No, you could have multiple.
The UPI was actually more sophisticated in the sense that you could have a UPI ID. So it could connect to a, it's called a virtual ID. It can connect to a bank account. You could use an account number.
You could use a mobile number as the address. You could use a virtual name as the address. So Nikhil at some access bank or something. So all these possibilities are there. So it just made it easy to use and real time.
So, and then two big things happened, right? So demonetization gave the first boost to... digital payments because people had to make payments and you know cash was not there and then the pandemic also people didn't want to hand over money they rather point to a QR code and send money so both these two tailwinds actually also drove the and of course the ease of use and the amazing benefits people got from it right so dpi journey let's say started in 2009 when you started building Aadhaar By 2013, you had largely solved for online authentication.
Yeah, already up and running. And benefit transfer via Aadhaar in a way. Yeah.
2013, eKYC came about. 2015, eSign also happened with DigiLocker. And 2016, you had these inflection points of demonetization, Jio launch, UPI, Bheem being launched. And Aadhaar reaching 1 billion. And Aadhaar reaching 1 billion.
At scale. At scale. You say this a lot.
Whenever you talk about a problem, you mention the word population scale. A lot. Like in every interview, I would have heard you say it two, three times.
That's why you shouldn't watch my interviews. Why does that word matter so much to you? Why does everything you build...
Because if you want to move the needle, and if you want to use digital technology to move the needle, you can only do it if you reach everyone. You know, when you build a private company, you build for 5, 10 million people, 15 million people, the top people, rich people, whatever. But in building a society scale transformation, you reach everyone. And India has a billion plus people.
So you have to think, how do you do things in a way that everybody can participate? How does everybody participate? So we have to give adha to everyone. We have to make bank accounts for everyone, mobile connections for everyone, you know, DBT for everyone. So the moment you want to do it for everyone, if you want inclusion to happen, you have to think at population scale.
They say when most people get very rich, very fast, they sometimes hate themselves and think the world is a terrible place and feel the world is unfair. When they get rich? When they get very rich.
Why is that? Psychology says that. I think it's a partly imposter complex, partly something. They feel that it's unfair that they got so much, whereas the rest of the world doesn't have. Oh, okay.
Post the success of Infosys, what drove you on this journey and is that why population scale matters? No, because at Infosys, I had done a set of things. But when I got a chance to do Aadhaar, I genuinely believed that digital technology used properly can make a big difference to lives of people. So I had that belief.
And it's there in my book, 2008. So that, and then I got a chance to practice those beliefs, right? I mean, it's not, you can have an idea. But somebody said, okay, fine, great idea.
Now go and do it, right? So it's a very different thing. So I got a chance to get it done. And then I realized that actually we have to do a number of things. So the last 15 years, we have built different parts of this infrastructure.
And I think it's made a difference because it's made it... more inclusive, everybody is in the system, it's helped to formalize, it's driving economic growth. All of that is happening because we have this way of thinking about digital infrastructure.
Right. And beyond 2016, when these inflection points happened, what came in 2017, 18, when did account aggregation come about? Interestingly, account aggregation, the first conversation also happened in 2016. Because if you look at the DigiLocker architecture, it was all about how documents could be stored, exchanged, and so on. So it was really a data exchange in some way, right? And then we wanted to implement that in different areas.
And we had an architecture defined for that. And it so happened that the Reserve Bank of India had come out with a... idea of account aggregator to for aggregating data.
Can you explain what is account aggregation? So account aggregation is, I as a consumer, let's say I'm a financial consumer. Is it mainly for financial data? No, it can be used for anything, but finance is the most obvious use case. So if I want to create a personal balance sheet of all my assets and liabilities, I have to get a bank statement from all my banks.
I have to get a detail of all my mutual fund assets. my insurance policies, my pension funds, my NPA, my EPFO, and bring all that together and construct my personal balance sheet. But how do you do that? So each of them has to give that data to you. So you need, again, an architecture protocol to say, go to my bank, get my bank statement.
Go to my mutual fund, get my mutual fund details, whatever. So that ability for individuals to construct their balance sheet out of all the various data, is what account aggregation is. But it has many uses.
So if I have to summarize account aggregation, should I say it's a repository of one's financial data? No, it's not a repository. It's just an exchange. The data resides where it is. It's a real-time system.
So let's say I want a loan. I go to my account aggregator who's working on my behalf. And there are many of them. I go to them and... And the account aggregator is not holding on to the data.
No, they don't touch the data. Right. They only, they're a conduit.
So I go to my account aggregator and say, get my bank statement from Bank X. So it's a real-time request. From me via the account aggregator to the bank, bank gives my statement for the last six months transactions or whatever. Then I say go to GST system and get my... And they give you that data without being able to read the data? Yeah, they encrypt it and give it.
Right. No, they can read it because it's from the bank knows what the data is. But the account aggregator does not. So when it comes to the account aggregator, it's encrypted and signed in a way that they can't peek into it. Right.
So they have to give it as is to wherever you want to send it. You start from a source and give it to a destination. The source is called financial information provider.
So bank or mutual fund. The destination is a financial information user. The guy who uses data.
So if it's a loan, as an example, I'll go to my bank, get my bank statement. I may get two, three bank statements from two, three different banks. I may get my income tax.
returns from the income tax system saying look I've filed my returns and paid TDS so that data. And I'll give all this to a lender and then the lender looks at that data and says yes this guy is eligible for a loan. And all this is happening in real time digitally. Same thing for a business.
A business can also get its bank statement, GST returns, give it to a lender and lender gives a loan. So it's actually enabling reducing friction for financial activities. Right. And this was conceived in 2016? Yeah.
And it got implemented when? Over the last seven years. Right.
And it's been rolled out. It's scaling up now. There's a group called Sahamati which is orchestrating this whole thing.
And hopefully Sahamati will become an SRO, self-regulatory organization. All the banks have tied up. Many lenders are using it.
So, but also it's not just, lending is just one activity. Lending is the bulk of it today. But personal finance management, people are building apps.
Other people are building capital market apps. So it's, again, SEBI is completely on board, RBI. So it's a movement to transform financial services. Right. What comes next in the India Stack DPI journey after account aggregation?
Well. A couple of other things which happened was the Fastag system, which I had designed in 2010. And that was to give everybody, every vehicle an ID. So just like Aadhaar gave every individual an ID, Fastag essentially gives every vehicle an ID, which is connected to a value instrument, a wallet or a bank account. And that streamlined the movement of vehicles on highways.
And today that does 2.5 billion transactions a year. Will it get to the point soon where you don't have to stop? Yeah, I mean, you don't, you have to slow down because it takes time for the RFID tag to be read.
But it's pretty efficient, I mean, compared. If you want to make it completely without stopping, maybe you have to use a different... But it was good, it reduced from people waiting for hours to seconds, so it's not bad.
So Fastag happened. Fastag was also a platform. Fastag was a platform where a vehicle could make a digital payment on the fly.
So the first use case of that was electronic tolls. But it could also be used for congestion charges or for parking. There are now startups that use it for parking. So again, a platform thinking applied to vehicles. I also wrote down IPO applications online, tax collection online, remittances, SIP plans, startups, digital IDs, digital rights.
For example, ASBA, which is when you apply for an IPO, you reserve some money for the IPO. The money is deducted from your account only if you get allocation in the IPO. That was completely made with...
Now half the transactions are IPO on the mobile phone, UPI, SBAR transactions. That made IPO so much easier to do. And now they're doing IPOs for secondary markets. So I think all, I'm sorry, using UPI for secondary market.
And then the whole, you know, over time Indian, everything goes to real time. So you know, T plus two trading, T plus one, T plus zero trading on the capital market side. Real-time bank account transfers, real-time, like for example today when Indians file income tax, the refunds come in a matter of days. Right.
Nowhere else you have this kind of stuff. Right. So now me as a 20-year-old or 25-year-old figured out the IndiaStack journey broadly. What do I build today?
That is your innovation. But give me some suggestions. I mean, look. Let's say I'm able to raise one crore rupees from somebody.
You give. What do we build on top of any of this today, which is like low-hanging fruit? Opportunities. Well, you know, it's up to people.
I mean, when I see the companies that use this, I see an idea that I would never have thought of. You know, the other day I met a company from Lucknow, which is helping roadside, you know, these street vendors to get supplies. So they're like a B2B company that provides all the supplies. For example, if I'm running a small stall, food stall, I make omelets or masala dosa or whatever. Earlier, the guy, and then many of them, right?
Earlier those guys would get up at 3 in the morning and go to the market and buy eggs and bread and spend 3-4 hours buying stuff, vegetables, whatever. Now these guys built an app for them to order and it's delivered to the cart. And they pay with UPI. So this could not have been possible before that. Suddenly that guy saves 3-4 hours a day and then he can have a longer day and he can have higher production.
And that's thousands of... street vendors are using this. Just, I mean, who would have thought about this, right?
Or there's like JAR, for example, which does daily savings. You know, every day it takes a little bit of money from your account. Do you think that's a good idea? Yeah, why not?
It was invented by Syndicate Bank, as you are aware. There are examples of similar companies in the West which round up and save. Yeah, but it's not on this, you can't, it doesn't do it on this scale.
See, in India, the Pygmy deposit. was invented by Syndicate Bank in the 1920s. So they had these pygmy deposit collectors who would go to everybody's house every day and take five rupees and accumulate in the bank account.
So it was invented here. A hundred years later, we have a completely digital version of that. They do 22 transactions per month per user.
That's a hell of a lot. So the fact is that in India, you can now build a system where I can set up an autopay and deduct. five rupees a day from my account and save it in the bank account or save it in a gold loan or whatever they do.
That is a remarkable way to bring everybody in. What is FIDE and the backend protocol? Yeah, so what happened was we were thinking about how to...
Three of you, right? Yeah, so we had worked on UPI, right? So Pramod, myself. We had all worked on UPI and we saw how a well-designed UPI protocol could unbundle payments. And suddenly everything changed.
So we said, if you can unbundle transactions, then you can create new business models where different people can talk to each other. Explain what you mean by unbundle. See, let's take e-commerce. How does e-commerce work today?
I go to some platform. I get an app from that platform. I look up some product on that app from one of the suppliers listed on that platform who has gone through the checks of that platform. Then I order something. And then their delivery agent delivers it to my house.
That's how, that's an integrated e-commerce. But suppose you take this e-commerce transaction and unbundle it, say. And this is like UPI thinking. UPI also, it's not everything, my account, your account, it's all different. So on my app, I can place an order from any supplier listed on the ONDC grid.
And I can use any delivery agent to have it delivered home. Would you say ONDC is something built, is Beckon a protocol on which ONDC is built? Beckon is a protocol which is a general purpose protocol. Explain what that means. It can be used for many applications.
It can be used for anything. It's a transaction protocol. It's a way to do transactions, to buy things.
And FIDE, at FIDE, you guys came up with... FIDE is a non-profit which stewards this protocol. This protocol is an open source. Just explain protocol in another way, right?
So everybody understands. Yeah. So it's, again, going back to what are the rules?
What is the language of a transaction between two parties? I need a way to describe that. So UPI described, as I explained earlier, a...
payment transaction between two parties. You're talking about the rules of the language. The actual, yeah, actually what are the things in it.
the structure of it, what are the fields and stuff. So when you say Beckon is a protocol, you mean it's a bunch of rules of the language Beckon in a way? No, it's a bunch of rules for transactional stuff, for people to buy and sell things. And it can be used in commerce.
So ONDC uses Beckon as an underlying protocol. And ONDC today does 12 billion transactions a month. It's used in mobility with Namayatri, which does 150,000 transactions a day, where people are ordering auto rickshaws.
Again, it's a different model than an aggregator because you pay directly to the taxi auto guy. And it's being used in energy. We are building something called unified energy interface.
People can buy and sell electricity. So anywhere that you have to buy and sell things between different people in a consistent way, Becken is the way to do it. Becken is a way to do it. So if I want to build something on top of Becken, how do I go about it? Well, you have to choose where you want to do it.
Do you want to do it in commerce, mobility, financial services? Let's say mobility. Let's say I want to build something using the transaction protocol Becken, something for like transport.
That's what Namayatri has done, right? Namayatri has built a platform where I can... Find an auto on Namayatri.
Exactly like I do it on my other platforms. But when that auto comes to me, I bilaterally make a transaction and pay the auto rickshaw. And I pay 100%. And I pay immediately. So this is different from being an aggregator which is sitting between you and the auto.
It's more of a discovery place. And the way the platform makes money is that... Auto rickshaw driver pays a daily SaaS fee to use the platform. So it's a different business model. When you say he pays a daily SaaS fee, what do you mean?
He pays the company 25 rupees a day to use that platform to get customers, unlimited customers for the day. So unlike in the other previous model, if I pay 100 rupees, then maybe 30 rupees goes to the platform in between. In this model, if I pay 100 rupees, 100 rupees goes to the...
driver of the auto. So me, if I'm a business called Namayatri or ONDC, I build on top of Beckin this transaction protocol as service and then I charge a SaaS fee to the end user who uses it. Does that work in reducing the monopolies and duopolies which exist?
It will open up the markets and create more opportunities. I mean, so it's the full scale of what you're saying what can guys do. Guys can disrupt many industries using this approach, which you could not do earlier. And you're going to start seeing that. I mean, I'm seeing some early signs.
See, the first use of these protocols was to mimic what's already happening, right? So, Nammeyathri already does what mobility aggregators do. Or ONDC does what e-commerce guys do.
But now... Energy was a new idea that you could now have a way to transact energy, like charging stations. I can go to a charging station and buy electricity through UAE. Or tomorrow somebody may say, I'll do a Beacon based version of quick commerce, where instead of my investing in dark stores, I will use the retailers in my neighborhood.
Right. And I'll provide a quick commerce capability. Is that a good business idea?
I don't know. I'm just, you're asking me ideas, no? I don't know.
You are the guy who's investing. What do you think of AI? I've heard you speak of AI in a manner where you firmly believe that large language models and building a model is going to be commoditized. It's already commoditized.
And nobody should spend energy on that. But you speak a lot for the Indian language and the nuance of it. building wrappers on top of these models to cater to that.
Can you elaborate on that? What do you think? Yeah. So there's a couple of, two, three different points here.
One is large language models, in my view, will become a commodity. They are already becoming a commodity. The drivers for the Western guys to spend billions of dollars is essentially because they're in a competitive land grab situation. So they all have to invest because not to invest is dangerous for them.
So they collectively are investing $200 billion a year, buying all chips, putting data centers, all that. But they have to do it because if they don't do it, they'll fall behind. So they have a compulsion to do it.
But the good news for us is all that is going to lower the cost and access to AI. over time. It's already happening. And then open source AI is coming.
People like Meta have just launched Lama 3 and they're going to give it away free. So I think this building LLM is a mugs game for us. It's great for somebody there but not for us. Our job is to leverage all this technology in a very low cost way to deliver value to our people.
Can you give us a couple of use cases? And Indian language is the use case which we have taken as one of the top priorities. Because in India, there are 22 official languages.
There are 300, 400, many more, thousands of languages, dialects, this, that. And if you really go back to population scale, as we talked about it, how does everybody interact with the computer? To use a keyboard, you need to know English.
Even to use a touchscreen, you can do visually, but you can't read. But if you can speak to the computer, and if you can speak in your language of choice, your mother tongue, then suddenly you can access the entire knowledge in the system. So it's a dramatic increase in access.
So if a farmer in a village in UP can speak in Hindi to the computer, and the computer can go and get some information on planting techniques to him and tell him in real time back. Suddenly we have made the world's knowledge accessible to that farmer in his or her language. So that's why AI is useful to bridge this gap of communication. So that's why language AI is important for India. I mean AI is going to be like electricity is everywhere.
So everything will have some AI component in. We are looking at how we can use this stuff to make people's lives better. Right.
Do you think voice will be big? I remember like when we were doing broking 10 years ago or 15 years ago, there were systems where you could say by voice, by particular stock, X quantity, Y quantity, all of that. It never really took off. But do you think voice-controlled automation is a big use case in the near future? Well, if you don't know how else to access, I mean, you may have a choice by either saying, buy X stock for Y rupees or by keying in, buy X stock for Y rupees.
We have a choice. But the guy who can't read or write, the only way is by talking. So Nandan, growing up, whenever I thought of artificial intelligence, AI per se, maybe I was wrong in thinking of AI as AGI, but I thought it's when somebody else can think for me.
Whichever model I've... kind of used a little bit of. It doesn't feel like somebody is thinking for me. It feels like somebody is executing better what I'm thinking.
Do you think it's right to call what we have today AI? Whatever. Those guys got together some 60 years back and called it AI.
We can debate whether it's the right name or wrong name. But I firmly believe what you said, which is this stuff is going to make our ability to do things better. Certainly not a replacement for what human beings can do.
And in terms of all this money being spent, like I track the American markets and I look at how much money these guys are spending on hardware. Yeah, over 200 billion across. But the revenue from AI is far lower than the capacity that people are building.
And to a large extent, one can assume this capacity will become, it'll depreciate very quickly in value over the next five years. So do you think we are in bubble territory because the money going in is significantly higher than the revenue coming out? No, at some point, point if the returns don't happen and markets may choose to correct.
I'm not a markets guy so I don't know about that. You decide that. But you have to look at the incentives for these guys. They can't afford not to do it. So they would rather spend 50 billion dollars a year each of them then not spend 50 billion dollars.
Maybe say 50 billion dollars. They make a lot of profits. But if there's a big platform shift in five years and they don't have it, then they're out of the game permanently, right? So the cost of not doing it is higher than the cost of doing it. So you have to think like that.
You're making a lot of investments in AI companies, right? I once got Tanuj in your house and I asked him, tell me all that you guys are doing. No, we do.
Most of this is philanthropic. I mean, I fund AI for Bharat at IIT Madras. These are mostly philanthropic. But are you seeing anyone in India do anything relevant? You can tell me India and outside of India, which are the companies which are doing the coolest things in AI, which have the biggest use case.
No, I think there are many. There's a company called Sarvam which is doing AI stuff. Bhavish is doing with Kruthim.
I haven't, I don't, I'm not really interested in the company so much. My interest is how do we use this stuff at population scale, low cost, to make a difference in education or health or farming or whatever. So what is the fint in it? Siddharth has been talking about this for years.
I think he told me first about this two years ago or three years ago. What is the Finternet and why is it so interesting as an evolution of whatever we have today? Well, in some sense, Finternet brings together the traditional, the mainstream financial system that we have and marries it to the possibilities that cryptographic techniques have taught us. You know, the concept of blockchains.
the fact that everything is on one chain, immutability of data and so on. So there are some very powerful ideas that the cryptography has brought in. What is Finternet? What is cryptography?
Cryptography is the fact that you used all this technology to make things immutable. In the sense you can... If you look at a blockchain, everybody can see the same thing. And every change made is also visible to everybody.
So it gives you the tools to create, instead of having a reconciliation between all of us, that acts as a public reconciliation of whatever it is, some transactions. So that's what is used by the cryptocurrency guys. So let's separate cryptography or crypto technology from cryptocurrency.
So that capability of immutability of creating chains was used to create Bitcoin and Ether and all these things. Basically everybody can see and verify the transactions. Without any, in the old world or in the current traditional world, you know there's a custodian, there's a settlement guy, there's a regulator. All they, the cryptocurrency guys came from this saying that we don't need authority. So when it, if you don't have authority, how do you resolve things?
We'll use software to resolve it. That's the broad theory, right? But there's a lot of good, very technology there. But that has gone to its own issues because when you start building financial systems that are not part of authority, then you end up with all the costs of that, you know, money laundering and terrorist financing and all that stuff.
And we saw what happened to many of these crypto guys. Can't that be scam? Like whenever I looked at the crypto world, if I were to...
Think regulation is everybody viewing transactions verified either by proof of work or proof of stake or even proof of time. If the chain in itself is not large enough, can't it be easily scammed or conned into going in a certain direction? I don't know enough to answer that.
But they took a technology, but on top of that, they layered. currency and they layered ideology. So they said we'll create our own currency. And they created some 15,000 currency. Everybody creates some coins, ICOs and all that.
And they said we don't want authority, we don't trust the state, so we'll do it through software. So that got them into all their issues. So what we have done is we said look, we should separate all this.
There is some great technology here which we can use. for high volume, low cost transactions. And if we can somehow bring that high volume, low cost transaction capability to the current world by a common way of doing it, then suddenly we can take our mainstream world and turbocharge it. That's exactly what Finternet does. So Finternet allows you to essentially airdrop like a modern engine into the existing system.
Which is interoperable. Which is interoperable. So if it's in the banking system, then I can take a bank deposit and following certain rules, I can tokenize it.
And then I can trade the tokens very fast. And then when I want the deposit back, I can convert it back to a deposit. So the ability to take different kinds of assets, convert them to tokens, then use a high speed engine to transact and then untokenize them at the end.
It's a common thing whether it's a bond or a stock or whatever. So that ability in a nutshell is what the internet does. It's an architecture to do that.
And when it's implemented, if it's implemented, what will be the first use cases of it? That depends. I mean, if it's in capital markets, it could be, you know, new types of ETFs.
Right. You know, which allow you to do, which allow you to create consumer products, which are different bundles of things, but you can trade them very fast. Within banking, it could be tokenized deposits. It could be enabling private assets to be, you know, collateralized.
I mean, we don't know exactly where this will go because we have given a framework and different sectors and different countries and different entrepreneurs and different regulators will identify different use cases. The interesting thing is this goes back to our thinking that if you unlock something, then innovation will flourish because you're defined the rules of the game. So that's what we're doing. I'm going to ask you some questions like really fast answer questions.
What do you think of CBDC? Where is it at? What is the future? CBDC is a central bank currency, but CBDC ultimately will be one more category in the Finternet.
So you need users. I think the big use of CBDC actually is in wholesale. So when you do cross-country transactions, you can do settlement much faster if you both use CBDCs.
Today's settlement complexities go away. Capital account convertibility. Do you think India will change?
Is it good that we are in the, like you mentioned, the LRS cap and all of that. Will that go away? I mean, again, it's not my decision, but I believe the trend is it'll go away.
In my view, it'll get more and more liberalized because India will get lots and lots of dollars. Only we get $108 billion of remittances, that'll go to say 200 billion. We'll get FDI.
And LRS outflow is maybe 20 billion. LRS is 20 billion. Inflow is 100 billion.
Just remittance. Forget about other things. So sooner or later, the system will have to liberalize.
So the way they'll do it is they'll do it progressively. So they'll take the LRS cap up or NRIs per year. Now they can take a million out.
So they'll keep on the edges, liberalizing it, in my view, over the next 20 years. Whether it'll be full counter-debilities or it's not. my decision.
It's a political decision. But do you think more will come in than go out? India is a high growth economy. The FDI will come, FII will come, remittances will come. And crypto?
What's your opinion on the crypto world? I think the crypto world has a lot of issues. And now we're seeing the cost of that. So I think that's where the Finternet is important. The Finternet essentially brings all the technology advantages of crypto world into the mainstream economy.
Over regulation, under regulation, what's happening in India today? It depends. I mean, I think some markets are well regulated, some maybe under, some over. I mean, there's nowhere in the world there's one perfect answer for that. Where do you think we're leaning on?
Which side? Some total? I think we are on par with I really don't know about overall regulation, but certainly on digital regulation, I think we are on par, better than most people. Socialism versus capitalism?
I mean, I'm a believer in markets and capitalism, but I also believe society must define rules for competition and innovation. And last words to a 20-year-old me starting work tomorrow with one crore that I borrowed from someone, what do I do? I think I've never seen so much opportunity. I think...
The ability for young people today to create successful companies has never been higher. But I'm confused with too much opportunity. I want you to tell me one.
No, that is, you have to, I mean, you have to, there's something called the tyranny of choice. You know, there's so many things to choose from. You go to a supermarket, there's 17 varieties of, you know, cheese or something. Then you get lost.
So half of life is doing, deciding what not to do. So you chip away and remove stuff from your thing and focus on a few things. But the opportunity is huge.
Sector-wise, any particular opinion? Like right now we're hearing a lot of defense is big, energy transition is big. Like I personally have been researching what to do in energy transition for a year.
Any outlook on that sector or any insight that you have? Energy transition is huge. It's going to involve trillions of dollars.
But where do investors make money in the private markets is not clear. Because energy transition is a big capital game, right? I mean, building massive solar plants or stuff like that.
So we have to figure out... See, there's two, there's big energy and little energy. Big energy is the grid.
Little energy is thousands of homes having batteries and rooftop solar. So the opportunity is actually in this little energy where there's lots of players and how do you make that efficient? How do you create markets there?
Right. Yeah, that's it. Thank you, Nandan. Thank you so much for taking the time to do this.
Thanks, Nikhil. I hope you didn't get bored. Not at all.
Very, very exciting. Okay, super. Thanks, guys.
Boss now, ho gaya. I wanna go. Why not? He's too easy.
He's too easy. He used to shoot it for 8 hours. Yeah.
Now he shoots for 7, he does. I know.