Subscribe
Share
Share
Embed
upGrad Enterprise aims to build the world’s largest GenAI learning initiative to enable high-growth companies to embrace technology’s transformative business impact. Hosted by Srikanth Iyengar, CEO, upGrad Enterprise, the GenAIrous Podcast, will curate an exciting roster of global experts and guests, who are at the cutting-edge of Generative AI, and its varied applications in the world of business.
Welcome to the GenAIrous Podcast, where we unravel the fascinating world of generative AI and its transformative impact on business globally. I'm your host, Srikanth Iyengar, CEO of upGrad Enterprise. At upGrad Enterprise, we're building the world's largest Gen AI learning initiative, empowering high growth companies to leverage cutting edge technology. Each week, join me and the roster of global experts as we explore innovations shaping the world of work as we know it. Let's get GenAIrous.
Srikanth Iyengar, CEO - upGrad Enterpris:Hi, everybody, and welcome to another episode of, the GenAIrous Podcast. This is our last episode of season 1. We've had a whole series of really interesting conversations, and, today's episode will absolutely not disappoint. My guest today is, Jo Aggarwal, a dear friend, the cofounder and CEO of Wysa. Jo is at the cusp of, AI today. Jo, welcome to the show.
Jo Aggarwal, Founder & CEO - Wysa:Thank you for having me, Srikanth. It's lovely to be here.
Srikanth Iyengar, CEO - upGrad Enterpris:Great to have you here. So, Jo, look, we're talking about AI and Gen AI. So why don't we just start with, AI? I mean, it's obviously something that's been in conversations, whether they be dinner table conversations or boardroom conversations for the last probably year and a half to 2 years. The newspapers, the magazines, the online, blogs are all filled with it. But how do we make sense of Gen AI? Is this, like, something radically new, or do you think it's a progression of past technologies? How do you see it?
Jo Aggarwal, Founder & CEO - Wysa:You know, Srikanth, I think it's it's it's a milestone, like, one that we have probably not seen in our generation, before. So I I do think that it's big. When, ChatGPT first came out, and it wasn't even as good as the work versions that we've had, that was just a year and a half ago. I remember going to Davos, and all the global CEOs and ministers and everybody had just one question, is it gonna take our jobs? And, there was a great sense of, you know, something big has happened in the even in the circles that were least tech literate.
Jo Aggarwal, Founder & CEO - Wysa:And there was also accompanying that big change, that recognition of big change. There was also a lot of fear. And since then, the world has gone through many, many hype cycles. But what I can see from our side is that literally you take out a product using one large language model, and there are 5 that are better than that by the end of the sprint. Like, in a 2 week sprint, you already got outpaced by technology.
Jo Aggarwal, Founder & CEO - Wysa:That's the pace at which it's growing. But the way to think about, let's just take a smaller, you know, a 100 year view, and then I'll take a 3,000 year view of this, so in terms of how to look at it. From a 100 year view, I would say that, generative AI you know, nobody thinks about the launch of personal computing as a milestone anymore. Right? We talk about the launch of, mobile or the launch of social media because that's what the next generation has maybe seen since 2012.
Jo Aggarwal, Founder & CEO - Wysa:But when we were growing up, it was, you know, personal computing came in from mainframes and changed the world.
Srikanth Iyengar, CEO - upGrad Enterpris:Yep.
Jo Aggarwal, Founder & CEO - Wysa:But really from the time of personal computing coming in, and there were series like Space 1999, which thought something big will happen. You know, there's always this been this promise of science fiction, the promise of computing that somehow science had has under delivered on until today. And I feel like generative AI is the first time science has delivered on the promise of computing. The other but, I have an interesting story for you because on a 3,000 year view, the last time, we had this kind of an inflection point, in my view, was, I don't know if many of you might have read, you know, about the genetics of India and how there was a time around 500 BC where we used to have a lot of mixed genes and then all of the castes started having very, clear segregations, intermarriages started stopped and so on and so forth. There's some inflection point at that time.
Jo Aggarwal, Founder & CEO - Wysa:At that time also was the launch of the first libraries. So at the Takshilas of the world, at the Nalandas of the world, you had the first libraries. And, this is a story about, Vedas and Brahmans, which you wouldn't expect in an AI podcast, but I'm going to go anyway. So when the library happened, the concept of not killing a Brahmin used to be a concept of not killing a library. Because a Brahmin literally lived his entire life to be the oral repository of all knowledge.
Jo Aggarwal, Founder & CEO - Wysa:The moment you had libraries in, in Northeast India, the moment you had Nalanda and Dakshala and you know, I think the first one, was Takshala, if I'm not mistaken. You had the concept of, will we become redundant, but at a very physical level, like, will we get killed? Because now who needs us? Mhmm. Because people can just pick up a book, and it's accessible.
Jo Aggarwal, Founder & CEO - Wysa:There were always scriptures, but they were not accessible to anybody. Mhmm. And that's the kind of moment we are seeing here where an entire set of people, an entire, you know, profession or profession after profession are feeling, that they may not be relevant at all. And what happened in my view down from this point, I'm going from historical fact to my hypothesis. Up until now was historical fact.
Jo Aggarwal, Founder & CEO - Wysa:I'm connecting the two dots of the genetics and this. And my hypothesis is that just like we're trying to build guardrails into AI, they build guardrails into our scriptures. So there is, there's many scriptures that don't have those guardrails, those who didn't feel threatened. Like, people who were writing about mats or Ayurveda don't have a lot of guardrails about not killing Brahmins. Or if you don't feed a Brahmin, this doesn't happen, that doesn't happen.
Jo Aggarwal, Founder & CEO - Wysa:There's a threat of your profession. There is a threat of your, you know, of your physical existence. Yeah. And all of that guardrails gets written into the scripture before it goes into the library. And I feel like we're at that point where not only are we building large language models, but the large language models in Gen AI that we built is going to represent all of our fears.
Srikanth Iyengar, CEO - upGrad Enterpris:I think the I a fantastic analogy, Jo. I think the 100 year of you, of course, I agree with you that today, you know, PCs or laptops are, you know, a dime a dozen as they say. But when we grew up and we went to business school together, we shared 1, you know, laptop each, you know, 5 or 10 of us, so I can imagine. But but the analogy for the 3,000 year analogy is fascinating. I mean, putting the getting and and for our listeners, there is no religious connotation here.
Srikanth Iyengar, CEO - upGrad Enterpris:But like you rightly said, the role of the human only gets amplified or accentuated. It doesn't go away. So completely completely agree with you. And and I look. I know that with Wysa, you're at the cusp of Gen AI because Wysa is an offering at the cusp of AI and, of course, mental health. So tell us a bit about Wysa and what you're doing differently as a consequence of Gen AI being on the, landscape.
Jo Aggarwal, Founder & CEO - Wysa:Absolutely. It's been very exciting for us at Wysa. So, first of all, you know, I always believed that the promise of computing was the ability to talk to an intelligence that wasn't yours. But until generative AI, we had to simulate that. So you had natural language understanding, which I come from like you.
Jo Aggarwal, Founder & CEO - Wysa:I come from the learning world, the elearning world. And, I used to get very frustrated that you couldn't change people's attitudes and behaviors with elearning unless you could listen to them. Because people change attitudes and behaviors when they feel safe, they feel heard, and they can experiment. Now, once natural language understanding came about, I said, okay. At least I can make them feel heard.
Jo Aggarwal, Founder & CEO - Wysa:Even if what the bot is saying is going to be based on a protocol that therapists have said, so what Wiser would do is make you feel safe and heard, hold space for you, understand, get you to really come out with your emotions, your thoughts, what the situations that you're dealing with, and validate them. Find all parts of you and just make you feel okay about them. And then help you reframe from, you know, the unsolvable, uncontrollable problems to the more controllable ones, from a way of looking at the world that feels bleak to a way that gives you more power and more control in that situation. And that's really what their therapists do as well. So that's what we were doing at Wysa up until GenAI.
Jo Aggarwal, Founder & CEO - Wysa:Now all of that works beautifully. About 470 people today have said that we've saved their life. 7 and a half 1,000,000 people have used Wiser globally in 95 countries. So we're at scale. We're used by all the major health systems in the world, major insurers in the world, large employers.
Jo Aggarwal, Founder & CEO - Wysa:So all of that's going really well. But every time we get someone asking us, can you do this in our language? We're like, this is another 3 years of work. And so language was a huge barrier for us. Voice was another huge barrier for us.
Jo Aggarwal, Founder & CEO - Wysa:All of that required clinical safety to be the 3 years that we spent building 1, and then it took us that much time to add a Spanish and then a Hindi. It was very, very laborious to recreate all of that. Today, we added low resource languages even like Marathi, and we've got a version of Wiza that can work in any language. So that kind of power generated AI gave us. But what it also did was it gave us our own moment of why should we exist.
Jo Aggarwal, Founder & CEO - Wysa:And, you know, the first question clients ask us is, why shouldn't I just talk to chat GPT or Gemini, why why would I talk to you? And in our own balanced scorecard strategy map, you know, we have added a bubble that says be better than GPT. Because even if we know the answer to that today, that may not be true tomorrow. So we literally have to add a strategy within Wiser to say that in all of the things that we do, we have to be better than a general large language model. But today, at least, Wysa has been able to bridge that gap, between what a GPT would do, use that, augment ourselves with large language models, but really do what we do really well, which is coach people.
Jo Aggarwal, Founder & CEO - Wysa:So, we become in, low resource settings. In rural Maharashtra, we become "Man ka coach, jo badle soch". You know? You take, somebody who can actually be met where they are even if they have no concept of, mental health. You become someone who can do this in 30 languages.
Jo Aggarwal, Founder & CEO - Wysa:We're now going to be launching very soon, open sourcing data to make sure large language models are safe because we figured that out in the last 8 years. So the other way it's changed us is that we've realized that, you know, people are going to talk to LLMs whether they're from mental health or not about their mental health, and they're not safe. And we may know some of them are safe, but we don't know all of them are safe in all of the languages. So we are going to add to the science of making sure everybody can put those guardrails in place and put that in the open source. So there are many ways in which you look at it.
Jo Aggarwal, Founder & CEO - Wysa:One, it opens new avenues for you, new ways for you to grow, but also it adds to your responsibility to make sure that the world moves forward in a safe way.
Srikanth Iyengar, CEO - upGrad Enterpris:No. Fantastic. And I love that line. For our non Indian listeners, it means the mind coach who helps change thinking or helps refine thinking. So just, you know, fascinating.
Srikanth Iyengar, CEO - upGrad Enterpris:And but let me come back to the sensitivity and the safety part. Right? I mean, mental health is still a taboo in many parts of the world. You know, people acknowledge it exists at a societal level, but when it comes to the individual, people are very private about it. Trust is a big issue, or a barrier or a bump that you've gotta cross.
Srikanth Iyengar, CEO - upGrad Enterpris:So how do you make something that's so sensitive? How do you make it safe, and how do you convince people that it's safe?
Jo Aggarwal, Founder & CEO - Wysa:So firstly, how do you make it safe? And 2, how do you make people feel it's safe? You know, there are almost 2 different strategies that you need. Making it safe means that, you, you make sure that even if the worst happens you assume the worst will happen. You make sure that even if the worst happens, people are safe.
Jo Aggarwal, Founder & CEO - Wysa:So you have to draw your lines very clearly. For us, that meant that even if we get hacked and everybody's data gets leaked and it gets put on the Internet, We don't have any identifiable data for anybody. So no matter what stories come out, nobody knows whose story it is. Mhmm. So Wiza was designed with a privacy first principle that said, yes.
Jo Aggarwal, Founder & CEO - Wysa:We will be end to end security, but nobody can guarantee that there won't be a hacker tomorrow that beats them. No one's immune to that. How do you peep make people feel they're safe? The interesting thing is, you know, the concept of safety comes from from a psychology perspective. Risk cues and safety cues, which are mirror neurons, are, are used to activate.
Jo Aggarwal, Founder & CEO - Wysa:So when you smile with your eyes, people get safety cues. And if you give a very strict expression, people will get risk cues. Right? So when you're talking to a real human being or you're hearing voice, your voice prompts, etcetera, your mirror neurons are picking up pauses, interpreting them, all of that in safety cues and rescues. And someone who's anxious is often picking up rescues so much that therapists really have to be trained to give enough safety cues for that person to open up, and that can take 2 or 3 sessions.
Jo Aggarwal, Founder & CEO - Wysa:Now we don't have mirror neurons for AI. So, research has shown, not just with us, but even with the VA in the US, that people open up to AI 2 to 3 times faster. That in Wysa, what takes somebody 3 sessions with a therapist to do, they will open up and do that work in 10 minutes and so. So that's the power of you already having safety cues with the chatbot because chatbots have historically maybe been incompetent but have not been judgmental or hurtful to us, while humans, we have a lot of stored trauma from.
Srikanth Iyengar, CEO - upGrad Enterpris:Let's, let's, talk about, you know, how AI, offerings are designed. You know, you also talked about the fact that increasingly, we will see, you know, public large language models being leveraged, but private LLMs being stacked up on them to create different use cases, if you call it that. And in some ways, Wiza is similar. But, you know, there are different approaches, you know, companies take, whether they be start ups or large companies. And I think many of them start with the technology and then get to what can it do for the user.
Srikanth Iyengar, CEO - upGrad Enterpris:But I believe, you know, you, like many others, also think that it's probably better to start with a human and then work backwards. So tell us a bit more about that.
Jo Aggarwal, Founder & CEO - Wysa:So I wouldn't say start with a human. I would say start with a problem. So, start with a problem and, you know, technology is an amplifier. It's a problem solver, but it isn't the solution looking for a problem. So you don't start with AI and say, oh my god.
Jo Aggarwal, Founder & CEO - Wysa:I had this fantastic experience with AI. I want to turn that into a product because other people would like it. And I'm hearing a lot of that happen. You know, I I got dating advice from it, which was great. Now I want to create a dating app because I think it can give great dating advice.
Jo Aggarwal, Founder & CEO - Wysa:A lot of that, you have 0, defensibility because ChatGPT or Gemini, you know, all of these are just going to get, Lama or Mistral. Whatever it is, just gonna get stronger. And, it's going to take all of your use cases. So your defensibility when you're starting to build something comes from a deep domain expertise in your area. Mhmm.
Jo Aggarwal, Founder & CEO - Wysa:And that depth of domain expertise in both understanding your users' lived experience and in understanding the domain. And that's where, Srikanth, what you said about start with a human comes in. So you start with a user, and you start with a domain, and you start with the stakeholders. So do I know how to make a doctor's life better or a therapist's life better? What they would need from a note taking app.
Jo Aggarwal, Founder & CEO - Wysa:Now, yes, there'll be lots of people doing note taking apps, and technology will probably become the least of their barriers because all of them will access the same elements, layer something of their own on top. They're no longer saying, oh, I have the coolest AI people because AI is now, you know, on tap. So now what do you do? Right? So what we need to start developing is that intelligence that our grandmothers had, that illiterate intelligence that just, you know, instead of saying go to market says, oh, there is a horse and there is a meadow, and I know how to get the horse to the meadow.
Jo Aggarwal, Founder & CEO - Wysa:Yeah. You know, that didn't require you to go to business school.
Srikanth Iyengar, CEO - upGrad Enterpris:Yeah.
Jo Aggarwal, Founder & CEO - Wysa:And, yeah, I can do everything else, but knowing how to get the horse to the meadow still requires illiterate intelligence. But somehow many of the MBAs still are missing.
Srikanth Iyengar, CEO - upGrad Enterpris:Yes. Yes. No. That that that that makes sense. And so let's let's shift a bit.
Srikanth Iyengar, CEO - upGrad Enterpris:We're talking about people who are adults today, who are at the workplace, but we now have a whole generation of, you know, kids, children. You know, you and I are both, we we are parents. We we've seen the you know, how how kids evolve, but, kids today are being born into a world with any eye. You know, it's it's it's an integral part of their life growing up from the day they are born. And, this is very different from kids who are born, let's say, 20 years ago into what you'd call a Google generation when when search was ubiquitous.
Srikanth Iyengar, CEO - upGrad Enterpris:And, you know, it tells you how old I am that we were grown up we were born in a world where email was just about gonna start. Right? I mean, let alone Google. So, obviously, it changes the way we we sort of, operate. So how do you see the impact on children who are coming to this world now in the last few years and going forward?
Jo Aggarwal, Founder & CEO - Wysa:It's an interesting question. There's a great book called the anxious generation, which I think is one of the definitive books written about, just the current generation that has come up. And, I feel, again, probably biased by my work in mental health, and the fact that I went there, with AI. But, I do feel that the definitive impact of technology, the way he puts it in that book is, that we overprotected the physical world. So there was no unsupervised play anymore.
Jo Aggarwal, Founder & CEO - Wysa:You know, we would go to school on our bicycles and have loads of time just to ourselves. Right? No no adult supervision, especially from the ages of 8 onwards, not even, you know, or maybe even younger. It wasn't, and now you have 16, 18 year olds being ferried around all the time. And so we've created this very overprotected physical space and zero protection virtual space for people.
Jo Aggarwal, Founder & CEO - Wysa:So adulthood is being experienced in a completely unprotected, quite brutal, very anxiety inducing virtual environment. So the whole virtual world that is coming. Now it's up to us. We're also, in parallel, bequeathing the next generation, possibly the most polarized world there is, there has ever been, where the polarizations have used to be at a cast level or at a tribe level. They have now seeped into our families and are probably going to seep into our brains.
Jo Aggarwal, Founder & CEO - Wysa:So we'll have polarized inner cells and polarized Thanksgiving dinners and, you know, and so on and so forth. So you're really talking about a generation that's walking in with that context. All technology amplifies that context. Right? The good thing about it is that the generation that you would call in the Google generation, I feel values authenticity.
Jo Aggarwal, Founder & CEO - Wysa:Is beginning to acknowledge. Like, we were beginning to acknowledge climate change, and we started giving them the tools to do something about it, and now they judge us for it. I think this new generation is beginning to acknowledge polarization as an issue, acknowledging that they need authenticity and real world experiences, acknowledging the dangers of the virtual environment. They will build the tools to fight that using Gen AI, I'm hoping. And I think the next generation that's just being born will use those tools to solve it.
Jo Aggarwal, Founder & CEO - Wysa:I'm very optimistic that that next generation is the one who's going to help us come back to the basics that are good, all that is good in the world. But I think the generation that is just coming into the workflows, the millennials, are the ones who build the tools, and Gen AI will help them do that. Will we amplify all that is bad in the world as well? Yes. But that is automatic.
Jo Aggarwal, Founder & CEO - Wysa:I think what is my call to action for the next generation? Use Gen AI to solve for polarization. Use Gen AI to make the virtual world safe. Use Gen AI to help people feel connected with themselves with, the world around them, and reduce their anxieties so that we can leave a world that we feel, you know, we are leaving a world better than the one we inherited.
Srikanth Iyengar, CEO - upGrad Enterpris:No. Absolutely. And and and there is hope. And and let me let me just ask one question, though. Obviously, technology amplifies the good and the bad.
Srikanth Iyengar, CEO - upGrad Enterpris:And one thing I keep hearing about from our clients and also our learners is biases. You know, we when somebody we don't know tells us something, we often have a healthy dose of, skepticism wherein human interaction. Mhmm. But when technology tells us something, whether it be a Google search or whether it be an AI model today, we trust it implicitly or most people do. And, you know, on one side, you've got lots of usage.
Srikanth Iyengar, CEO - upGrad Enterpris:Other side, you have synthetic data being ad reaching models all over the place, and that only amplifies biases. How do you see that playing out? Because, you know, data can now be manufactured almost infinitely to support a model.
Jo Aggarwal, Founder & CEO - Wysa:Mhmm. I trust human intelligence in that. I trust that, you know, for instance, we call it WhatsApp University. Right? We know how much fake news there is on WhatsApp.
Jo Aggarwal, Founder & CEO - Wysa:So when somebody says I got this as a WhatsApp forward, we snigger. So, that you know, that's WhatsApp University. So that's clearly most likely to be fake news rather than true news. You know, when we used to read The Onion, we knew that was satire even though we trusted what's put online. But, and in the same way, on social media, people are becoming very aware of the credibility.
Jo Aggarwal, Founder & CEO - Wysa:And this next generation has a is seeking authenticity. I think a lot of the ability of fake news to influence the older generation is higher because they haven't grown up with those filters. So when they see the written word, their biases, their bias checker doesn't come in. Our bias checker comes in. My son's bias checker is immediate.
Jo Aggarwal, Founder & CEO - Wysa:And I am quite confident that human intelligence will adapt.
Srikanth Iyengar, CEO - upGrad Enterpris:No. Makes sense. So, Jo, this has been fascinating. We've gone, you know, 3000 years in the past. We're in the present.
Srikanth Iyengar, CEO - upGrad Enterpris:We've talked about the future. Thank you so much for joining us today. You know, we've we've covered a whole host of things. And as always, very, very insightful and, inspirational as well. So thank you very much.
Jo Aggarwal, Founder & CEO - Wysa:Thank you, Srikanth. And it's always a pleasure to talk to you.
Srikanth Iyengar, CEO - upGrad Enterpris:And that concludes season 1 of the GenAIrous Podcast. Just to recap, we started in early July, and over the last 3 or so months and 15 episodes, we've covered the diverse and transformative impact of Gen AI— Enterprise transformation, legal and ethical implications, its impact on the environment, the effect on mental health, how academia changes, what various industries are doing to embrace this inflection point, and so much more. A big thank you to our guests who joined us from all over the world, San Francisco, New York, London, Singapore, and India. Thanks again to Shantha and Nitin for making magic happen behind the scenes.
Srikanth Iyengar, CEO - upGrad Enterpris:Most importantly, a special thank you to you, our listeners, for tuning in. We'll be back soon with season 2. In the meanwhile, don't forget to subscribe to GenAIrous wherever you listen to your podcasts.