Listen to the full episode below:
Cutting Edge startups in fortune 500 all in one room talking about how emerging technologies are changing our world.
Swish: Today I'm here with Jonathan Perez and Karthik Balakrishnan. Jonathon is the Director of Global Innovation and FinTech Partnerships at ScotiaBank, on his day-to-day he focuses on coming up with innovative solutions to meet the upcoming needs of customers. Jonathon, welcome to the show.
Jonathon: Thanks for having us.
Swish: Karthik on the other side is the CTO of Wysdom.AI. Wysdom.AI provides conversational software that helps enterprises gain insight into their customer journey. Welcome Karthik to the show.
Karthik: Thank you so much, pleasure to be here.
Swish: So Jonathon, just to define the key terms of the episode at the large, FinTech is quite a buzzword. How would you define FinTech, what does it encompass?
Jonathon: So in my team, we define FinTech as, yes, financial technology companies, but we don't focus on companies that are already well established, like the traditional big vendors that the bank already deals with. We look at FinTechs as emerging companies in the financial technology space, not only for financial products, but also enterprise solutions that will help the bank basically deliver products and services faster.
Swish: And then Karthik, conversational AI, how do you break that down?
Karthik: Every decade or so there's a distinct shift in the human machine interaction paradigm. Right at the start of the internet era, businesses went “why do I need a website?” and then everyone needed a website. And there was the mobile revolution, every business that had a website went, “do I need a mobile app?” Or “why do I need a presence on a mobile phone” and then the mobile revolution happened. And conversational AI is one of those paradigm shifts, where the new interface of human machine interaction is going to be driven by conversation.
And we are just about at the tip of the iceberg, I would even argue it's just about getting started. And aside from the aspects of human machine interaction, the vast majority of information in enterprises today postulated to be 80 to 90% is unstructured. And not only is it unstructured, it's textual, and we as humans have codified this in language. So one of the ways to actually unleash that information and start making sense of it in a way that it can be actioned and derive some benefit out of it is again, conversational AI. So conversational AI is really the entire spectrum of applications that deal with the ability to process unstructured information, and then perform certain actions.
Today, I can't wait to sit down with Jonathon and Karthik so I can ask them some questions I've had on my mind for a while now, should job loss be a concern when it comes to conversational AI? How can we better protect customer data? And finally, will customers want to talk to an AI over real humans? Personally, in the short run, I do believe that AI will lead to job loss as lower skilled laborers are going to have their task automated. But in the long run, just like I've seen, society evolves. As society realizes that lower skilled laborers need higher education to be able to get jobs, our system is going to adapt.
And in the long run, I can fully anticipate the jobs of tomorrow to look very different from the jobs of today. I want to dive deeper into the integration of conversational AI into consumer facing positions and what impact that leaves.
Swish: What are some misconceptions both of you experience when it comes to what you do? Karthik, let's start off with you.
Karthik: In terms of people getting it I think for the most part, people are very intrigued. And things like Black Mirror and all the sci-fi episodes come to mind. You know personally, I'm very impressed by how some of these sitcoms have portrayed technology and the future of it. It's very thought provoking. People do get it. I think to me, the most telling was that building something using generated AI via this bot had actually learned to talk back like a customer service rep. And it was actually generating answers on the fly on its own. And like, just having looked at some past data, it was able to come up with new answers on its own, and I'm showing that to my mom. And then her first question was, aren't you going to be taking jobs away? So I think people don't get it.
Jonathon: No, I wouldn't say that’s the same with my mom actually, I think my mom still thinks the same thing that she thought when I started a Computer Science program for my undergrad and she still thinks that I do tech support. Which is fine. I try to explain it to her and she still picks up the phone and calls me when something happens with our computer.
Swish: I love that in terms of the day to day, Jonathon for you, why do you think it's so important to have consistent communication with your customers, especially when introducing new technology?
Jonathon: I would say it's important for two main reasons, one for adoption and for feedback, when you're launching a new technology, you need to constantly talk to them for what they need to use it for, why they need to leverage it, why they need to jump straight into like changing the way they do things today. Because normally if I ask you today, there's something that you do, it's working for you, but now I'm saying there's a better way for you to do it. If you're one of the pioneers, then you would say yes, I'll definitely jump into it, but if you're like the other 90% of the people, they won't do it. So there's an important component for transitioning them from how we used to do it, or even if we used to do it before, to how we want you to do it now.
So that’s one component. And then the second one is, if you truly want to know whether what you're launching is working or not, you just need direct communication with them. Super important not only for improving your products and services, but also to know you’re prioritizing the right things at the right time. Sometimes the solutions and products you launch are the right ones, it just happened that the timing wasn't right. I would say for us, it's important for those two elements. The way we do it is there's a major Customer Experience Program at the bank, which is actually a global program, it’s standardized across all our different markets in which we look at how are the products that we're delivering working with the customers? Are they delivering the right value? Are we adding the right features? And we even take some of the information they gave us, and we use it to prioritize the roadmap that we have.
Swish: And Karthik, you guys obviously represent a new technology. Like you said, it's becoming more and more mainstream every day. Do you feel like you guys are going through and seeing that companies that are adopting conversational interfaces are empowering their customers? Are they just pushing their customers into one direction and trying to get them to adopt it?
Karthik: I think it's a little bit of both. And I would agree with Jonathon first and foremost, timing is important. You can't push customers beyond the vote. I think a great example of that was I think it was almost a decade back when Bell Canada had this conversational AI bot called Emily and it failed. Because it simply couldn't deal with the variety of accents that you get in Canada, right? It wasn't good for its time. So you can't force people to use something against their will. So that was the failure right there. But then flash forward now, I think as time has evolved, I think it's very important to realize that we are all being subtly manipulated by the big tech companies right from Google Search to the Alexa's of the world that start making conversations cool. We see so many conversations between humans and humans, humans and bots, tons and tons of data. It's incredible how forgiving people are when bots make mistakes. People laugh and joke about it, they wouldn't be half as forgiving of a human. But it's people's willingness to find something cool and be willing to go with it and companies are leveraging those signals, as these technologies mature, I would say the cost of adoption starts going down.
Companies thought to put it out there, let people come to it, and gradually as people started using it organically, it becomes the new normal.
Swish: And then in regards to job loss, obviously a very key topic associated with AI. Jonathon's mom thinks that he's working tech support and in the future, I bet a lot of people within that industry might be scared by conversational AI in terms of taking away their job and taking away human decision making.
How would you respond to that Karthik?
Karthik: I look at it with caution. I wouldn't stand up and say oh, jobs are going to just go, I think it's going to happen little by little. We humans are incredible at creating problems. We think we solve problems and then we create multiple other problems. It's a good argument to have how much grief emails give you, now you can get messages across in a second but the expectation is that you also reply within an hour and then people get fatigued and that creates a whole new industry of people dealing with digital addictions and so on.
So in terms of jobs, I would say we've constantly had jobs shifting, I think my view is two things will happen. One is the entry level for entry level jobs will be much higher, like the skill level needed for an entry level job will be much higher. So you wouldn't be able to just walk into a place and you know, start a mundane job. A second aspect is it's going to happen organically, we do have time to adapt. When I say it's going to happen organically, it doesn't like you know, one day, suddenly half of the people are going to get fired. It's just that positions are going to get harder and harder. But who knows, society will adapt as we start building these things.
There are going to be several other industries that open. You ask an interesting question about does it take away human judgment? I would definitely say there's some early evidence suggesting that it takes away some reasoning, it takes away some common sense, maybe seeing a lot of that like the amount of spams and forwards that people just hit a button and send it to all their friends and relatives without really vetting the source of validity of some of these things that they send. So critical thinking judgment.
Jonathon: Unfortunately, my hypothesis is that we'll see a little decline, probably an increase with depression rates as well or mental health issues. Again, early evidence suggests some of these things are imminent.
There's a report that the World Economic Forum issued, I believe last year or towards the end of 2018. They mentioned that basically 100 and 30 million jobs will be created in the next four or five years in which the top two are AI machine learning. So, yes, it's removing a lot of the non judgmental jobs. But I would agree that the skills to get into a new job are definitely higher now, in terms of whether we're seeing that the decisions are being removed, the easiest issues, I think they will be removed. Just think about yourself right now, if I have to drive from home to my office, I just trust whatever Google Maps tells me. I don't even try to think about how maybe this is a different way, although my wife would say something different. But for the most part, I trust what Google Maps is telling me, but we're seeing a lot of calls or even ask from our customers in terms of advice, so when they want to plan for their future. I wouldn't say that in the next two years, that's gonna totally change. There's still going to be some ask and spend some time understanding who I am and what I need, and when I need things.
I find that in banking, specifically, there's one of the most important things right now is we can bet coffee or know when you're going to need certain things. But overall, that doesn't mean that you specifically are going to need those things at that time. So if we know you better, if we know when you're going to need each of the products that we can offer you or the services that we could offer you. Again, it's all about timing. So that element will definitely be part of the future.
Swish: And for banks, in particular, customers need to trust the programs that they're working with. How do you guys ensure privacy and security when rolling out new technologies?
Jonathon: I would say that it's a number one priority for Scotia Bank. And it's actually something that goes from the C-Suite to our entry level employees. You have to go through privacy training, you have to go through security training, a lot of new applications that are monitoring what you're doing, which some people might say is good or bad, but it's just something that we have to take care of. Because if we have a data breach, it's gonna affect our reputation, it's going to affect who we serve, how we serve them, and it's definitely gonna have an impact on our business. So if we don't take care of our data, which is our main asset, then it's one of those things that could put you in big trouble.
Swish: And then for Karthik, conversational AI tools like Alexa and Google Home, they've all been scrutinized for listening to conversations when they weren't activated or turned on. And obviously, it's a breach of privacy. But an argument that can be made is that way Alexa can become better for you to personalize everything that they're offering back to you in terms of suggestions by listening to conversations that are happening throughout the day. How do you feel like the balance between privacy and quality should be made in conversational AI?
Karthik: First and foremost, I do want to talk a bit about one of the biggest myths, a lot of the media goes out and makes it appear like everything in AI is self learning, and there's a big bad brain that can just automatically understand everything. It's far from true. The vast majority of how these smart engines if you may call it that the Alexa’s ,Google Home, Siri’s or voice assistance, it's actually a bunch of humans training it behind the scenes. Yes, they do need training, yes, they will increasingly keep getting smarter. But I think thanks to the Facebook breach last year, Alexa, Siri, Google, they all came out and they confessed that they actually have 10s of thousands of people listening to all of these conversations. And it's people behind the scenes sitting and like optimizing these models. So a fact in AI is that it's not one and done. Once you adopt AI, it's like getting a plant, you've constantly got to give it sunlight, you've got to water it or else is going to wither away and die.
In terms of privacy and this justification, should you give more of your data, and you'll get better service. I believe people usually vote with their choices. If people are paranoid people wouldn't do it. And if people are not paranoid people would do it. And that's something that each and every one has to decide for themselves, their comfort levels. There are people today who don't use credit cards, right? They're simply paranoid about that information, getting digitized and getting out there. So I think bots, majority of people, like I mentioned earlier are increasingly finding this as threatening, but at the same time they're willing to give up their data for better service.
Jonathon: It's just about the communication. I think people don't want to know that their information is being captured when you haven't told them. Yes, if you told them already, they might say, you know what, I'm okay. But if I find out that you're actually capturing more of my data, I wasn't aware of it, then that’s when it becomes a problem. But if you have that communication with the customer, and you tell them, hey, just gonna get this piece of information to make this better for you. We have seen good traction with that. When I think about myself, when I go into Amazon, and I know that some of the things that I'm being offered, I’m like, why am I even seeing this thing? And I'm not even close to buying any of these things. Then you wonder, is it because they don't have information to know what I'm looking for to just predict what I'm going to buy? What's going on here?
Karthik: Sometimes it's an AB test. You simply don't know you're being tested to see if you'll click. And you mentioned an interesting point, I think the whole book of consent management passed some distance ago but we are seeing the earliest days with regulations like GDPR.
Our hypothesis is that in the future people will have process consent management companies if you can call it that. For example, I may decide that Scotiabank is my identity broker, Scotiabank will be responsible for the identity to everyone from the Alexa to Google homes of the world to the Home Depot's and for me, I can go to one place, I trust Scotiabank, I can go authorize to get my permissions, what aspects of my information are they allowed to get? And in one click, I can revoke? And I think we are getting there little by little. I think right now, in the name of speed and innovation. A lot of that has been overlooked. But it's going to be a matter of time before people get more conscious about that. And these kinds of trusted identity brokers come into this ecosystem. And then it's question of like you mentioned very rightly, Jonathon, concept as far as people know what's being used, how it's being used, they can be party to it, and you need to give them the ability to revoke it at an instant of a notice, right shouldn't say well, it's going to take another month or so.
Jonathon: You should be in a position as a responsible organization to delete any information that the customer doesn't want you to retain. And I think that's going cross industry. I think right now you're seeing the first steps in the UK, it's mostly within the banking ecosystem, you can basically pull your information from one bank to the other, but think about what's happening in Australia. It's actually cross industry. So if I just want to open a new hydro account, I can just use my information that I have in my Australian bank and just go and try and open an account so I think there's definitely going to be that portion of but it's kind of gonna be delayed entity which this is all your information and you can port it from one place to the other from even if you want to buy it on Amazon or whatever you want to do after the information has already been validated for you.
Swish: And then Karthik, there are thousands of ways, especially today for a consumer to interact with the brand. How do you guys at Wysdom take all of that disparate data, put it all together and tell a brand, this is what customers are saying right now, this is what they’re ooking for?
Karthik: I'm gonna say that's going to happen in three phases. Currently, we're seeing the earliest days of that, where if you look at the vast majority of conversational AI deployments today, they are depersonalized. It's a one size fits all, you go to any bot today is going to give identical answers no matter what your past history is with the company. And what the company's strategy is, we are moving to the next phase where we are now building AI that will help this AI get better. So here's the AI that’s serving up customer experiences. And here's another AI that's taking everything into account, everything from the customer’s spouse history to the customer's level of understanding of the issue, but it's very important that you are able to talk to someone in a way that they understand and of course, the organization strategy on hand.
So if you would want to upsell something to someone, how do you generate more revenue out of this customer? For some organizations, it's a good strategy to let a customer go right at that point, how do you lead that customer churn, so having an organization strategy into account is very important and actually brings a fully intelligent bot, if you can put it that way that's really aware of the past history strategy and cohesively binds all that together to sell a customer experience.
The third phase is this AI that is currently being built that optimizes the application that’s serving the customer will eventually be embedded within the application that's serving the customer. So literally think of a bot going out there, fielding a day long of questions, coming back and telling the human hey, these are all the things that need to be fixed, or here's how your process could improve and here's the net gain for you. Or you need to go create these kinds of digital assets to serve this kind of a requirement.
Swish: And for Jonathon for Scotiabank. What does the ideal customer journey tend to look like? And how do you feel like we could use AI to help guide a customer but not overwhelm them?
Jonathon: There are two types of customers who we usually get, so they're the ones that are looking for a solution and the ones that are not looking for Scotia, and somehow you need to make yourself appealing for them. For us, there's a big push to compare the economics of the digital customer and the non-digital customer, and I think that just speaks by itself, the fact that the numbers are way better when they come digitally.
The one thing that we're trying to solve for right now is how does the ecosystem of a customer look like? So when you think about a customer, right now, banks are offering just a financial services product, which is a transactional relationship whenever you need something, I'm here for you. But if you think about banks, they were not thinking, Karthik needs a house, let me help him find a place to live, the way we used to think about it is, hey, here's a mortgage, let me sell you the mortgage. And that's not the case anymore. Right now, it's, you know what, actually, if you think about the people that are moving to Canada, if you think that you want to start a small business, there are many examples in which let me just have the person or the business throughout their value chain. And by the way, at the end of the day, here I am, I have all these products that can support you at these different points in the journey.
I think that change of mentality is creating a different customer journey than the way we used to have it before. Because like I said, we're an inflection point in which a previous customer journey used to be, let me just be there whenever they need the transaction, I'm here to execute. And the good example is the auto ecosystem through the dealers, you can get a loan for your auto. But what if I'm there for you to just help you set up the appointments to start looking at the cars, let me just help you throughout the process, it will become natural for them to work with you instead of just being there for the transaction.
Swish: And then Karthik, do you believe that when a individual customer is trying to interact with a brand and an AI appears, and they're conversing with a chatbot that it can lead to a disconnect because a lot of customers feel like especially you know, whenever I've used the chatbot, I realize I'm using a chatbot, you know, they don't talk like humans do. Is the goal by the way for them to eventually talk like humans or will I always know it's an AI and you just have to run with it?
Karthik: One of the most profound things that I came across as the best technologies out there are the ones that become transparent, you don't even know it's there. And AI is going a similar route. I think the end goal is to have very human-like conversations, you know, you don't have to talk in a command or control structure, you can just have freeform conversation. There’s a lot of challenges to getting there. I'll give you an example, you could be talking about a mortgage. And then I can just say, oh, what's the rate for that? So all of a sudden, now the bot needs to understand what that means. That's very easy for a human to understand but for a bot that's incredibly challenging or sometimes it's the absence of real well, common sense.
For example, so if I tell you something like the trophy didn't fit in the brown bag, because it was too big, so what was too big? And these are incredibly hard for a bot to do, because remember, we're approaching this not from a language understanding standpoint. Yes, on the surface, it's called natural language understanding. But really behind the scenes, it's a lot of math and stats, right? That's building that kind of understanding. So we as humans have evolved very, very nuanced and complex patterns in language. And consider that we've been around for roughly 300,000 years as Sapiens, for the bots, they've been around probably in the last five years, so it's very early days very, very early days.
And side by side. There’s also an adjacent to AI called ontology knowledge graphs, you know, which is adding structured information for these bots to be able to automatically reason and understand real world facts or arrange real world facts in a hierarchy. For example, if I tell you I had pizza for lunch, you know that I didn't make a very healthy choice, maybe then you may refine your question and ask me, oh, wait a minute, was it white crust or whole wheat crust? If I say whole wheat crust, maybe a healthier choice. If I say white crust, next question could be, oh, did it have cheese? And then you'll be like, bad choice. Yeah, you're able to make a judgement. You can make inference because you know what a pizza considers. But if you've never heard of Perogies before and I tell you, I had Perogies for lunch, you can make a judgement, right? And that's in a way how these bots are reasoning today.
Then working through this, there was a profound moment I realized that the truth is not objective. It's contextual, without context of the issue on hand and without context of someone's past truth, it’s different for each and every one of us.
So we've heard how brands are taking care of customer data, how humans are interacting with AI and whether or not using an AI in customer service will cause problems for brands.
But now I want to bring it back to the human level, and find how brands are incentivizing customers to not only provide feedback, but also give their data. A brand needs to be very transparent when telling their customers how they're using data. So it would be amazing if every brand had a portal, where after you've given your data to that brand, you can see exactly where it's being used and for what. The way Trufan gathers data is off the backs of players like Instagram and Twitter where we directly connect with their API. So currently, we do not provide any incentives to try to get consumers to give data.
Swish: Before we dive in. I want to quickly hear how Scotiabank deciphers what technologies are going to be the most useful for customers.
Jonathon: My personal view is that I try to stay away from, let me see what technology I want to test, and just pick what are the right problems that you want to solve. And then the technology's just the means to solve that problem.
So if you look at how we structure our partnerships in the ecosystem, we work with venture capitalists, we have academia, we have strategic investors, we have FinTechs. Like there are many, many ways in which we get the knowledge of what's happening in the industry. But I would say it's not about what technology we choose. It's about what's the problem that we're trying to solve. And then which is the right means to solve it.
So for example, the conversation on AI is if we want to get away from having in-branch transactions, and we know that there are certain things that people are coming to the branches for because they're going to travel and now they just need to activate a credit card, like this used to happen before. So if we have that data, we don't necessarily have to say, oh, I'm going to use this technology for enabling that. We'd say, what's the right way in which we could enable this and conversation that could be that mechanism?
Karthik: I couldn't agree more. You've got to always customize your problem statement first before you jump to adopt the technology, because at the end of the day, every technology out there, the reason it gets adopted is some sort of a return on investment. The classic example I can think of is Uber and Airbnb, it wasn't even about a decade back, people would say, don't talk to strangers. And then what happened? People started picking up strangers in their personal cars, people started inviting strangers, not just from your neighbourhood, from all over the world into their house. There's a very simple reason, money, right? And money is the single biggest motivator and then in line with money being the greatest motivator.
Swish: And then Karthik, when I look back at some of the reasons why I for example, completed surveys, I completed them mainly because I would get a gift card, right? Are these incentives necessary to gain a higher amount of customer data? And if not, how do you get customers to communicate directly with their brand in a more organic way?
Karthik: So there are two ways, one is what we see is having been done conversationally and via live with almost about 15, large tier enterprises. What we see is about on an average 10% of people actually leave feedback. So on one hand, that's in some ways sufficient because of the sheer volume, so you kind of know what to go do because feedback is important.
There's another aspect we're working on as a company, which is really to understand the, like you said, organic, you don't have to engage them. For the first one, I would say, yes, you need to roll out some incentives. Now, if you want to increase the 10%, to maybe 20 30%, and get more numbers, you've got to put some incentive out there because at the end of the day, customers are giving you their time. And you know, there's an interesting question like, if I asked you to just give me one minute, will you. If I asked you to give me 10 minutes, will you? You have some expectations out of that minute and out of those 10 minutes? Why would you give that to me, you're expecting some return and feedback is very similar. Now, having said that, what we as a company are working on is a hypothesis called the feedback pyramid, just think of a pyramid and at the very bottom is face to face communication and you communicate face to face. You have an enormous amount of feedback. You don't need to tell me whether you understand, I know chances are I can guess very well that you understood that you're paying attention to me, I can look at your body language, I can look at defensive body languages and I can kind of get a sense for where things are going.
The second one above that is, think of a voice call. The moment you make a phone call to someone, you don't see them. They could be saying “mhmm” but you simply don't know if they're actually listening to you. They could be busy on their laptops doing something else and multitasking, you lose some feedback. Now take that another step forward. And at the top of the pyramid, you have text. So imagine texting someone, how much feedback can you really gather? Almost nothing. And you know, going back to your earlier question in the first half, which is about bots being bad today, that's a challenge we're dealing with is simply relying on pure text, you know, 10% of people may actually leave feedback, but the other 90% we simply don't know. And when people do provide explicit feedback, a lot of us would do it for the incentive. But a lot of times people are more likely to provide negative feedback than positive feedback. And when they provide negative feedback, even if the feedback is explicit, it's very hard to understand sometimes whether the feedback is attributed to them having a bad day, to a dislike of the interaction they had, or to your brand.
Jonathon: The one thing that I would just add is, I would separate the incentive for feedback versus the incentive for adoption. If you're thinking about incentive for adoption, then your product might not be solving what it actually needs to solve. I always think about Uber and myself using Uber when I get the $4 or $5 discount, but not when I don't have it. Is it actually solving a real problem for me, probably not, I live a few minutes away from work, so I could walk, I could take the streetcar, it's easy for me to go to work. But if I have the discount, and it's more of the same, sure, why not? It's convenient.
So the incentive for adoption, it's something that I wouldn't recommend, and if you have to jump into that, then you just need to think about it. If you're thinking about the incentive for feedback, I would totally agree with Karthik, I think if you want to increase the population that is giving you feedback, for sure. I always think about my wife. She has a quarterly subscription to a company that delivers a monthly box with... items. But she continuously gives them feedback on what she likes, what she doesn’t like, and she does it very organically.
Swish: And that's really interesting because I feel like most of the time, I don't trust restaurant reviews, or even movie reviews rarely, because I feel like people that go online, take the time and effort to go and give a restaurant review normally are charged with negative energy, but restaurant gave you the, you know, the Pad Thai that you wanted, like, I'm not gonna go out and be like, great! But if they give you a really shitty Pad Thai, you might actually take some time out to be like, don't come to the restaurant. So I find that interesting. How can we take in more of those instances of trying to organically charge a customer to just provide positive feedback, because at the end of the day, if they know eventually that if I get feedback on the items coming to me, maybe the next batch will be more personalized to me, that could be really neat too.
Jonathon: Wysdom as an example, if the user bought, the one thing that you see at the end is was this helpful? and you could take two seconds to actually say yes, it was helpful. And we can try to dig deeper into why it was helpful? But what we need to know is, was it really helpful? And it's thumbs up thumbs down. It's that simple is that?
Throughout this episode, Jonathon and Karthik really opened my eyes to how AI is being used for brands and the customer first approach that they'll take. Just hearing Karthik and Jonathon got me really excited about the future of AI. I think specifically conversational AI is very interesting. The ability for consumers to get to their endpoint a lot quicker and more efficiently by going through an AI that recognizes what they need and has learnt from their previous interactions. That's something that we currently don't have with customer representatives or customer service that is very human driven.
I absolutely do think that AI is going to become more seamlessly integrated into our day to day lives, especially right now where everyone's quarantined, you're in your home and you're looking for better options, whether it's to get food, get your deliveries or meet people, it's more likely that AI is going to serve as a function to replace human interaction, but to more importantly, make your ability to interact with your community a lot easier.
Swish: We're getting into our final segment here, the rapid fire round.
I am going to individually ask you guys some questions, they're going to be very quick questions, hopefully some quick answers back and ideally, the answers are going to be full of fire.
All right, Jonathon, first question. What do you want to see invented in the next 10 years, get creative.
Jonathon: I’d like to see more clean energy and renewable energies solutions implemented, if you look at what's happening in Australia, I think that speaks for itself.
Swish: Number two, what technology should be feared?
Jonathon: I think it could be used for really, really good things in every single industry that it can imagine, but it can play it against us.
Swish: Number three, what is your definition of success?
Jonathon: The ability to influence people to do good for our society, and I totally resonate with it if I also am balancing that with having time with my family.
Swish: What is the best thing about working at Scotiabank?
Jonathon: Coming from Latin America, I can tell you that diversity is rare and at Scotia, it's unique.
Swish: Alright, Karthik, first, the same as Jonathon, what would you like to see invented in the next 10 years?
Karthik: So funny enough, it's also along the clean tech area for me. Honestly, I get very flustered trying to recycle, my whole heart I want to recycle and I find it in in some ways bewildering that we're building technology to go to Mars but we still don't have something that can solve recycling, at least tell me which bin to put it in, like I show something on the right bin and it lights up.
Swish: Second question, what technology should be feared?
Karthik: I'll go over the cliche, I'll say AI. To me, Black Mirror has been incredibly revealing on what some of these technologies can do. And we often think about utopia. It's very important to think dystopian as well, because you know, the threats are pretty real and I think I would say my favorite episode is Be Right Back, it's where a partner passes away. And this woman actually then starts uploading his text messages. And then they build a bot for her. She starts chatting with it. She then gives them some voice transcripts. And then it starts talking to her in his voice. So next thing she uploads pitches, and the company keeps incentivizing her better service. Why don't you upload some pictures? So she does that next thing, it acquires memories and it starts talking about those places as well. And events that they did together. The final step, they actually sent her a little MOS that in 24 hours transforms into the actual person. And you know, some of it while exaggerated, especially probably the mass converting into a body. I think some of the other threats are pretty real.
Swish: Number three, what is your definition of success?
Karthik: Success is basically overcoming adversity. I think life isn't the Olympics and not everyone runs the same distance.
Swish: What is the best and worst thing about being an entrepreneur?
Karthik: I would probably start with the worst. Very Honestly, I don't find anything bad, right? Yeah, you do have stress and it's not easy because you're literally pushing the boundaries, right? As a startup, you're the one who's constantly having to innovate, because that's your lifeblood. If you don't do it, you die.
I think that is in as much the best part about it as well as while you're doing that you essentially are in the realm of self realization, and just you know, those little things that come out of that that eventually add up to being something big is one of the happiest moments.
You guys have been listening to The Tech Haus podcast where we bring cutting edge startups and Fortune 500s to the table to talk about their contrasting views on how tech is changing our world. Stay tuned for our next episode. This is your host Swish, signing off.
And now let's hear from the sponsors that made this podcast happen.
Origins Media Haus is a boutique content marketing agency that specializes in creating high quality podcasts and videos for your brand. They've done it for so many people across Toronto and across the GTA. It's been incredible to be able to work with three really hard working women that know exactly where the world of podcasting is going and can situate your brand perfectly for it.
Trufan is a social intelligence platform empowering businesses of all sizes to make smarter marketing decisions. Use our platform Social Rank to identify your valuable followers and understand what resonates with your social audience all through one platform. If you're interested. Check us out at Trufan.io.
Free for 2 hours of content per month
$12 for 3 hours per month
$18+ for 6 hours and up
$5/month for Monthly Storage 50mb
Unlimited audio package: $9/month
Monthly Storage: Unlimited