Mini: From History to Hashtags: Shifting Sands of Engagement ft. Arvinder Gujral

very small content which will be labeled as made by humans which become very precious that hey this is actually made by humans so you will create a premium on that content but there are still a lot of startups in India you only look at India as a market and or look at us as a market came in and wrote this entire thesis that you know what we are going to fire most of our customer support 50% of customer support is being fired looking at customer experience holistically today how do you think that will evolve in the future [Music]

this episode is brought to you by Super Checkout, the world's first AI powered checkout that lets you split, combine, or co-ay with friends and family in real time. Whether it's shopping, gifting, or travel bookings. Now, everyone can chip in with ease. Check out smarter with super checkout. The Hustle Group Company. Let me introduce you to the Hustle Group Company. It's a lifestyle brand that's redefining streetear through the power of fashion, self-belief, and resilience. The second one being Discover Dollar, which is an AIdriven tech company that helps brands

and retailers recover hidden dollars from overp payments and leakages. Docs now intelligent platform empowers businesses of all sizes to rapidly collect, manage, govern, and collaborate on the data front, transforming your documents and making sure there's an impact on the business bottom line all in a secure and a single environment. Hi everyone, good morning. I'm here in Singapore uh and welcome to another very very short segment. I have someone very interested in me who has redefined a lot of different perspectives and created impact in the space of conversation,

social media, fields of consumer experience and so on and so forth. She's an incredible leader and I'm very glad. Thank you so much for joining me today and giving me time to thank you. Thanks for having me. Really appreciate your time and thanks for coming down to Singapore. Thank you. It's my pleasure. It's a fantastic city and I wanted to have this chat with you for a lot of wisdom and GAN and of course you're recording a small bite of a lot of GAN that you've already given me. I wanted to give this make sure that you know this goes out to

a lot of people out there as well and they get to learn from you as well. Uh for people who do not know Arvinda has been the ex managing director of Southeast Asia for Twitter. He was the first employee of Twitter back in the day and uh in in India and uh he's been able to scale it up across AP pack right so fantastic to meet you uh bunch of things right we start we'll start with social media considering that you know things are changing quite significantly over the last few years I wanted to talk to you a little bit about that today in today's day and age with the

evolving AI generative AI and so on and so forth it's almost impossible to distinguish between human generated and AI generated content I mean if you do a touring test I'm sure it's going go way past even with you know discerning eyes you're not able to differentiate between that content so this five quick question that I wanted to ask you the first one would be how has social media evolved with AI and how safe is it anymore okay so let me take it in two parts so how has social media evolved and safety of social media right

let me take the second part first cuz that's easy the safety of social media is not depend on AI the safety of social media depend on the people who create content and people who consume content right that will not change no irrespective of where the content came from whether the content was created by a human being or created by a human being using AI right the ethos of safety of the content because if someone has a bad intent right

they could have created that content even without AI because fake news is not new they did not need AI to proliferate across countries across geographies across across political spe spe spe spe spe spe spe spe spe spe spe spe spe spe spe spe spe spe spe spectrums you have been consumed and created on fake news forever right in fact if you Watch a documentary. Fake news was even used during World War II, World War I, uh toppling of democracies by creating fake narratives and it has

happened in kingdoms in wars that have been fought before. So, you know, AI can only, I would say, accelerate creation of content which may not be kosher, right? But bad content or good content will always come. The only thing AI will do is only accelerate the amount of content being created. Makes sense? And then it comes to know what how will it change social media. It'll change social media to the extent that the amount of content that we consuming now we will be cons thrown a

lot more content to consume. What happens when you have to read one book in a day you will read it. When you have read 10 books in a day you will skim it. When you read 10 50 books in a day you will want somebody to give you a summary of it. When you have read 100 books in a day you will say somebody give me one one bullet point for each book so I know what the book's ethos is all about. Makes a lot of sense. This is what's happening in social media until what

will happen with the amount of content that will quad multiffold because AI will help you generate more you as a consumer will have shorter attention span. So that's a second order effect, right? We anyways have what is a proverbial uh attention span of less than a goldfish which is 6second. Can you imagine now us having attention span of 4 seconds or 3 seconds cuz we've gone down from movies to episodes from episodes to clips on social media on Insta and others and we are finding even those large enough that

I can't watch the whole clip. Can I even make it shorter? Right. Very true. And so this is an unfortunate being because what you do is long- form conversations right of an hour plus right which will become a rarity which in itself become a niche which only a few will want to experience right and so social media will become and I see you know if I were to take the arc of history where social media or content

will evolve is there will be a lot of content majority of it will be created by AI right using human prompts Right? And they'll come to AI prompts creating AI content. Makes sense. A small minority over increasing time will be made by humans. So now you might now you will have labor saying content used AI made by AI. Right? Over a period of time and let's say 20 years from now you'll have very small

content which will be labeled as made by humans which become very precious that hey this is actually made by humans. So you will create a premium on that content. So humanled experiences, humanled content will become a premium. Mhm. And then content which is long for which again a premium that somebody has the audacity to spend money and invest time and effort and money and resources to create something long form knowing that majority of the content users will not want to consume it. Makes a lot of sense. So that I think

will create this dumbbell approach with one dumbbell being extremely heavy which is a short form content made by AI and another end of the dumbbell which is extremely small but more premium which will balance the premiumness will balance the weight of the dumbbell otherwise you'll have a very skewed dumbbell which is AI generate content being so heavy that nobody cares about the short form or the the long form content which is very short and that's fantastic right because uh this is this is giving us a glimpse into the future of how conversations will evolve

how interactions will evolve on social media platforms and beyond. on you know recently there was operation sindur that happened uh between India and Pakistan right here watched sort of counter strikes and in that in one of the summarizing points that was later debriefed uh it was mentioned that about 15% of the time of the armed forces went into tackling fake narratives which was like wow I mean for me I mean the world has shifted so significantly from you know a leader from armed forces coming on to debrief and stating a point like this Which means that AI would again be

utilized to distinguish between those two data points and very clearly labeling out that this is AI generated so that you know those fake narratives can be handled much better. I'm assuming that you know there's a lot of work going on in that space as well. Some of the startups that I know have started to tackle this to create those ethical AI uh guardrails as well. But coming to the next point right uh uh you know you've you've dealt with a lot of customer experiences you know across your journey over the years right over the last couple of decades or more. So in that

journey and looking at customer experience holistically today, how do you think that will evolve in the future? What's the future of customer experience going to look like? So let me So today I'm working with a startup called Enable X. We we in fact work with large brands and enterprises including startups to create customer journeys. Right now what we are increasingly finding is a lot of the customers and enterprises are coming to us saying hey I want to use AI.

Right? It's a very generic statement. Nobody knows what that means because the flavor of the month, but it's a true flavor. It's not I don't believe that AI is ephemeral like a lot of other trends that came and went. I think it's here to stay. It's just that the interpretation of what it means for you as a customer, as an enterprise or as a brand and what it means for your end customer that that interpretation will keep evolving and changing and like let me exemplify I don't want to give

motherhood statements. You are a large e-commerce platform, right? You have customers pinging you your customer desk every day. Where's my order? Do you have this SKU of this size of this color? Blah blah blah. You have so many questions. This is current customers. You have a lot of prospects coming to your website saying hey I want to know if blah. Today what will happen is you had you used to have a lot of humans answering them and mass. Then you say you know what I can create an AI chatbot that can answer all the questions. Great example Clara. Clara

came in and wrote this entire thesis that you know what we going to fire most of our customer support. 50% of customer support is being fired. All we have evolved a native AI stack that will solve for all these problems. We've connected disparate data sources because data is the is the key foundation on which AI sits. We have done all the hard work everything. The whole headlines came in all over the world. Customer support is dead. Teleerformance kind of companies will be over and done with blah blah blah blah blah. Less than a year later they've come back saying

we're rehiring half of those people we fired because you know it's really not working out because people want to talk to people. Exactly. Essentially what that tells you is human nature will remain the same. End of the day when and when I have to outsource a decision to you. I cannot let the AI be accountable. I want to choke some human's neck. You said this to me. I want to hold you accountable because you cannot dissociate yourself say oh my agent said this to you right

and not me. No your agent is created by your company. You're accountable for the answer they gave. If the answer is wrong I'm holding you as a human. So when strategic decisions are taken right a human has to be in the loop makes a lot of sense because humans will always trust another human there will be a mild distrust which may be high distrust today which because AI will evolve when you compare you will still have a mild level of distrust going forward and you will know has a

human being actually blessed this answer makes a lot of sense maybe not all the answers but the important ones you want someone to say this is what we believe is right now I'll give it further when you like take an example today you're talking to your bank, you go to the website and you chat with the AI chatbot. You get some answers from them. You say, "No, not yet. I need to know more." Right? At some point, the AI will say, "I don't know the answer to this." Right?

He the AI will have to get a human in the loop. that used to be 50%, it's becoming 30%, may become 20%, but there will be a maybe 10, maybe 4%, maybe 2%. But there will be a percentage in that CX journey where a human will be in the loop, right? When you have a premium member, let's say you're a you're a large fivestar hotel and you have Hilton honors member, right? You will not tell your Hilton honors member to go and talk to an AI chatbot.

You want to give them customer experience. You say, "Dude, I'm here for you. I will answer any question you have in real time and I will solve it for you. I will not tell an AI chatbot to solve it for you. Makes a lot of sense. So there will always be a segment of customers. There will always be a percentage of customer queries that an AI bot may or may not be able to solve. And even if they solve it, the customer will want to know whether a human has blessed the answer because your output of the AI bot is an input to

my decision to something else. If there is an error rate here, that error rate compounds. Imagine one AI chatbot gives an error rate of 0.1 gives an input to the other AI agent which then becomes 0.5 which becomes an input to a third AI agent because 10 like 1%. You can see the compound impact of that one error answer. Exactly. And that you will not know till the output finally comes to the end customer who say dude this is not what I wanted at all. Y this is the Chinese this is the Chinese

whisper right one small error compounds and you have some other answer at the end and this is the danger of AI and while AI is improving can we say AI is 100% right in 100% of the cases not in the foreseeable future I can't for cast the future because if I could I would be a very rich person today no fair that's amazing in fact love the whole compounding perspective as well Because majority of these algorithms are selfing and selfolving mechanisms, right? And by the time you realize the circle has been completed, it's already

spent so much of time in relearning from existing answers which it perceived to be correct. Yeah. And then the whole basis of the next set of questions and answers gets skewed as well again. Right. So fantastic point on the future of customer experience and the fact that uh that you highlighted is that trust premiumness of conversations and empathy human empathy will always be more premium as an experience to any kind of a customer. So that's why you always need a customer a human in the loop for customer interactions for premiumness to the to that conversation

to bring that in. Fantastic. very quick another one right uh there are you know you have been at the you know you you've been involved in the India Singapore corridor for a while now and you've been a prime example of someone who has built out an ecosystem in both the ecosystems right build your own credibility in both the ecosystems a lot of uh entrepreneurs from India want to be in Singapore and vice versa as well to tap into the Indian market what is your perspective or uh you know suggestions to or feedback to them so one of the things you know what I see Indian founders may

or may not a lot of them have already realized but there are still a lot of startups in India who create a uh who only look at India as a market and or look at us as a market right and my suggestion to them is while that is good also look at Southeast Asia not because I'm sitting here and I'm telling this to you I'm going to give you a very interesting fact India is multiple states right multiple languages multiple context and cultures so it's not a uniform country

right If you want to do a B2C play, you have to slightly, you know, make it market specific to make it relevant culturally. If it's a B2C play, like even if it's social media, right? Right. If you're doing an enterprise play, maybe it's uniform and you just target the top thousand enterprise customers. But if it's slightly more B2C in nature, you need to know where are the premium customers in India sitting and they're sitting in X number of cities and it's a X number of households

who can spend Y dollars a month which is actually a very small number which we have validated across you can look at any VC tech in India now they have come across India is not that it's a large country with a lot of people but the number of people who can spend $10,000 a year on discretionary spend is much smaller number right you take that to Southeast Asia and you take that percentage age of people who can spend the same amount of money in Thailand, Philippines, Malaysia, Indonesia, Singapore and you

will come to a number which is close to India and it's only five markets. Think of it as five different states of India. Right. Exactly. And like oh I can actually create another India by tapping into Southeast Asia. Obviously it is slightly more complex than India because now you have different countries and hence different regulations etc. So market entry has to be thought through for each market. Right. But it is closer to India.

Makes a lot of sense. There's a lot of cultural connect between the two markets in Southeast Asia and India. You will find a lot of Indians living and working in these markets as well. So you for you to get and it is much cheaper than going to the US. So for you to make a market entry and test the hypothesis of your product is very viable. Makes a lot of sense. So I would say give it a thought. Don't skip it because you think of India, South Asia as oh Thailand is for you know fun and frolic. Uh Indonesia

who knows what it is. you know, Philippines, you know, it's a cheap it's a poor country compared to India. Singapore extremely expensive. I don't know and it's very small population. If you have that generalization, I think that's probably the wrong way of looking at this market. You're typ casting it too much, right? To be honest, you know, and but that is the general perception. If you're a young founder and you you're starting something, you will not think of SCA, right? Makes a lot of sense. So that

that's amazing perspective. So people who are listening to him know this for a fact that there's a lot of scope out here and you can relook at these five markets as a mini India of sorts or equivalent to India itself but different regulations and but it's very solable right and huge marketing to as well uh thank you so much for this very very quick question any person living or deceased who you take inspiration from a massive inspiration or most inspiration from so you know I one of the things that I've been asked this questioning who are

your heroes in life And it's a very difficult question to answer because your heroes change, right? I love to read. And so one thing I realized is the people I appreciate more are the people who help me understand things which are completely out of my realm of understanding. Hence it is deep science. I'm not a physicist or a mathematician, right? I'm a data business guy like you and solving very simple problem that they are solving. M

so Richard Fineman is I would call him a hero or an inspiration like the things that he has you know how he has done in life and the way he explains things that part I love KL Sean big fan amazing big fan right he has brought the universe and the because if KL Sean did not only just talk about the explaining in English what science and astronomy means he obviously dumbed it down to people like even I can understand but for me the question is

once he explains it to The second order effect of that is real. You realize your place on earth and then you realize earth's place in the universe and then in the galaxy and this in the universe like where we sit like you're a pale blue dot which what the famous quote by call you're insignificant. Yeah. So you know if you go to connect disparate conversations you may have had or desperate things you've read these people inspire me to think out of the box and out of my what I naturally would have thought of. So

makes a lot of sense. Yeah. Yeah. And in fact uh fantastic to you know bring humility and perspective into place right so thank you so much for sharing that one last question for you you've also been consuming a lot of podcasts right a lot of content uh can you talk about maybe one two or three top podcasts or top content forms that you really like so in our conversation I said like the Indian podcast that I really like is the seen and the unseen by Amit Whma I love how it's very very long format so kudos and patience and

bladder to him to six six hours with somebody else. I'm sure they're taking a lot of breaks and editing it, but six hours to have a conversation with somebody uh is is requires energy. Uh and how deep they go. Uh and the how thoughtprovoking those conversations are is is amazing. I love history. I mean, I realized when I was a kid, I hated history cuz it was ro learning. Tell me this date what happened. I'm like, you're not really teaching me anything. You're just asking me to read and learn

something. But now I've been consuming a lot more on history. So there's a a podcast called the great dictator. Okay. Oh, the the greatest dictator something like that. I'll give you the name. And it goes into idiamin, Hitler, Stalin. Uh I'm just listening one on Shawara and uh so it's just interesting to understand where they came. All of them start with okay who were they as a child? Why did they become the person they became? And hence why so when you see we know what they did

but why did they do that make a lot of sense that's more important because what they did is the way we were taught history right he did this at this time right but I don't care I want to know why he ended up being Hitler what was exactly going on what made him Hitler there has to be some context to someone becoming and thinking that way that he's right because in his life in his mind he's right

right in our minds with with the in hindsight he was wrong right and he was evil but in his A lot of people thought he's right. People voted him. So why? Why did that happen? And if you can understand the why, you can see today's right-wing movement in various part of the world. You can understand the context of why it's happening again. Why did the genocide happen? Then you will say, "Wow, it's so terrible what he did, right? Never happened

again. How can humans allow it to happen? Dude, it's happening again in our time. What are we doing about it?" But why is it happening again? And essentially the tables are reversed. The people who were who were castigated by Hitler are the ones who are probably perpetrating some of it depending on your political spectrum you'll say they're also doing something similar. Another sector again for me history podcasts are the ones which I find grippy a lot makes a lot of sense. History repeats itself and you did mention this in our conversation

beyond the recording as well about intrinsic motivations. So this this podcast I'm assuming also gives you a perspective motivation behind a lot of actions that have been taken the so-called leaders who instigate such actions right so gives you a lot of perspective thank you so much that's that's phenomenal I'm going to go and check that out because that's very kind of very very inspiring for me as well to get into the human emotions human interest in motivations and understand how things happen the way they happen so thank you so much again for taking your

time and I'm looking forward to hosting you in India as well over a full full-fledged podcast I think like a deeper conversation exactly Thanks for having me. Appreciate it. Thank you so much. Signing off from City of Thank you guys. Bye. Bye. [Music]