What's Up with Tech?
Tech Transformation with Evan Kirstel: A podcast exploring the latest trends and innovations in the tech industry, and how businesses can leverage them for growth, diving into the world of B2B, discussing strategies, trends, and sharing insights from industry leaders!
With over three decades in telecom and IT, I've mastered the art of transforming social media into a dynamic platform for audience engagement, community building, and establishing thought leadership. My approach isn't about personal brand promotion but about delivering educational and informative content to cultivate a sustainable, long-term business presence. I am the leading content creator in areas like Enterprise AI, UCaaS, CPaaS, CCaaS, Cloud, Telecom, 5G and more!
What's Up with Tech?
Humanizing Technology: Omilia's Journey in the Fusion of Conversational AI with Customer Service
Interested in being a guest? Email us at admin@evankirstel.com
Step into the future of customer service with us as Dimitris Vassos from Omilia pulls back the curtain on how conversational AI is adding a human touch to technology. Discover the strides this two-decade-old company has made in not just executing tasks but in truly engaging customers, all while navigating the twisty road of generative AI, from combating fraud to the eerie realism of deepfakes. It's a behind-the-scenes tour of the revolutionary dance between automated interactions and the human agents who are finding renewed purpose in their elevated roles.
This Monday, get ready for an episode that's not your average tech talk. We're merging insights with real-world experience to reveal how AI and human customer service reps are crafting a renaissance in job satisfaction and customer interaction. From the peaks of voice biometrics safeguarding retail in its busiest seasons to my own adventures with AI tools that craft content with near-human flair, join us for a conversation that's as enlightening as it is practical, showcasing the seamless magic AI brings to the customer service experience.
More at https://linktr.ee/EvanKirstel
Hey everybody, happy Monday. Monday morning here in Boston. You know where I wish I were today, on a Monday morning. It's in Greece, where we have Dimitris from Amelia. We're going to talk all things conversational AI and beyond. Dimitris, how are you? I'm doing great. Evan, how are you Well? Thanks for being here. You're such an innovator in the conversational AI space. I have questions around this space. It's a fascinating area moving so quickly. Before that, maybe introduce yourself a little bit about your journey to start Amelia and your current mission.
Speaker 2:Absolutely so. I started almost 25 years ago. I'm founder and CEO of Amelia, so I've been with Amelia for 20 something years since I founded it, and O'Neill is a conversational AI company. We're transforming customer service. I started from IBM in the UK and back then AI wasn't even a buzzword back in 1997. And I got busy with trying to deliver automated customer service in a way that consumers would not hate, right, at least, which was the situation back then. Since then, I think Amelia has grown to be a leader in conversational AI. We're delivering natural language self-service solutions to customer contact centers all around the world across financial services, retail, insurance, utilities, automotive, any real really enterprise segment where customer service is strong, segment where customer service is strong. We're now taking this technology into other customer service environments, which is quite interesting the work we're doing lately with drive-through automation in the quick serve restaurant in the US. So that's also something to watch out for as we are going to be deploying across the nation. We're already doing so.
Speaker 1:Amazing.
Speaker 2:Yeah, that's really a game changer there, enabled by generative AI, of course, and the latest technology. So that's pretty much Amelia. We're all about restoring the human touch in automated interactions. We're all about elevating the role of customer service representatives to do more meaningful tasks than just, you know, answering a thousand phone calls a day and giving out balance inquiries or trivial requests like that. And another thing that is unique, I think, to Amelia is that we're helping customers fight fraud in the contact center, and that's become even more relevant these days with generative AI and the way that that technology is being used to create deepfakes, right, deepfakes for video, deepfakes for voice. So authenticating securely consumers in customer service has become even more a difficult task and a demanding task.
Speaker 1:Wow, so much to unpack there, so much interesting stuff. I'm glad we have you for the next half hour. Well, talk about one thing. You mentioned customer service. That humans don't hate. There's a lot of that going around, and how is conversational AI fixing that, changing that? I mean, it's been around for a few years now, but it seems only now have we reached a tipping point where you know the industry is sort of adopting the latest technologies and tools.
Speaker 2:Well, that's right. I think there is an inflection point where you know it used to be that consumers had to adapt to the technology to learn how to work with the technology, and, you know, whoever was not tech savvy would just be afraid of dealing with those automated systems and reject them outright. And the inflection point came, I think, with conversational AI technologies being so good at understanding natural language that humans no longer had to adapt to the technology. They could just be speaking to the system in exactly the same way as they're speaking to another human, and the technology would actually understand and not start saying this very, I think, common in. You can see it everywhere in YouTube, all these little videos of bad system implementations where it says I'm sorry, I didn't get that. Can you please repeat, right? So we are now in the era of generative AI and certainly the technology is absolutely up there, capable of carrying out a conversation with a human consumer.
Speaker 1:Yeah, that's quite a sea change. And what makes your technology stand out? It's a very crowded market. I saw a market map once with 3,000 different players and different niches and segments around conversational AI, but you seem to have some really special sauce. Tell us more.
Speaker 2:Well, it's a very crowded landscape indeed, very noisy. The technology is so accessible at first glance to so many people, to a broad base, from the large enterprises down to individuals who you know they decide to take this generative AI from Microsoft, openai, whoever and have a go at it. You know, with a chatbot and of course you know it's very easy to do a proof of concept and to display some sort of intelligence, but it takes a lot of experience to actually deliver enterprise quality results at scale. And this is really what Amelia is all about. We've been doing this for more than 20 years, even before generative AI existed, and we know enterprise customer service. We know how to safely deploy this technology so that we can both have happy consumers but also happy enterprises that are confident they are secure and confident that they get business benefits out of the technology.
Speaker 2:And you know we've seen some recent examples of chatbot failures in the news right, they revealed the larger legal implications behind using AI-powered chatbots from inexperienced providers where they just put a chatbot on a website or on a mobile app. And I'm particularly thinking of a recent event with an airline, I think, in Canada, where you know the chatbot provided an inaccurate response to the consumer, saying that you know they were willing to reimburse the consumer and that policy simply didn't exist. It was a hallucination from the chatbot, probably trying to be very polite and customer friendly, but it got the enterprise into trouble, right Legal trouble. So these are again. There's still a lot that needs to be developed and guardrailed around this technology before it becomes safe, and that's where the competitive landscape, I think, begins to clear up. There's companies that know how to do this safely, and then there's everybody else that's having a go at it.
Speaker 1:Great point. Well, you're a real practitioner and a pro your team, so that's great feedback. There's a lot of uncertainty and fear around AI replacing jobs and when and which jobs and how many and there are a lot of unanswered questions there, a lot of prognostication, let's say. Even last week, the head of an Indian software services company, one of the biggest in the world, said contact centers will disappear within a year in their traditional form, with human beings. But what do you think? What say you from your unique perspective on the industry?
Speaker 2:I always like to take the middle way. I think that, definitely, contact centers are not going to disappear, but I also definitely think that there's going to be a very significant impact of this technology in customer service. There's going to be a very significant impact of this technology in customer service. Ai, for sure, is poised to enhance human roles in various sectors, not just customer service. It's going to bring a new dimension to how we perceive and execute our jobs.
Speaker 2:It's a tool, tool and, in the context of contact centers, ai is definitely revolutionizing the traditional role of a human, of a human agent, who have often been engaged in very repetitive and mundane tasks and often also requested by enterprises to merely read of a script, of a script on the display, right on the monitor. So, you know, that kind of, I think, behavior we're going to see be alleviated by the use of technology and we're going to see humans, I think human agents in the contact center take a lot more meaningful role without the pressure to perform in terms of speed, which is, you know, a lot of the KPIs in contact centers today is, you know, speed how many calls can you get rid of right in a day? Calls can you get rid of right in a day, and that, of course, goes against customer service, goes against the user experience, goes against the agent satisfaction themselves. So we do expect agents. Once they start using this technology, they will be a lot happier agents and I'm pretty sure that we will have customer service agents for many years to come.
Speaker 1:Yeah, interesting to watch, we will see. And speaking of AI enhancing, augmenting or replacing humans even, what about your development team? I mean you have a large team around the world, including in Greece. How are you using that to either augment or scale your developer resources?
Speaker 2:Well, we're using generative AI in pretty much every aspect of our operations. One big aspect where we're using generative AI is in analytics. I think generative AI has been very helpful in automating a lot of manual tasks that we used to do to analyze our customers' applications and how well they're performing. So generative AI is a very good tool which accelerates, that, automates, that. It's still not replacing people 100%, but you can think of it as one person cloning themselves into times 100, right For certain tasks, not for the whole job role, but for certain tasks that they do within their day. So it is really exploding the productivity of our employees in certain areas. It's not that you can apply it horizontally everywhere, but that's, I think, a process where a lot of enterprises and a lot of companies similar to Amelia are currently engaged in is to find new ways of working, given that disruptive technology that has come about.
Speaker 1:Yeah, such an important topic. We'll be watching that as well. In the beginning intro you mentioned using voice biometrics to prevent fraud and you know, voice recognition, voice biometrics even have been around for a long time, but not the most useful technology. But what's changed? Because this has kind of come around again with a new opportunity with Gen AI, has it not?
Speaker 2:So what's changed, I think, is that the attacks have become a lot more serious now, like fraud and trying to fool the system. Serious now like fraud and trying to fool the system. And these attacks are not just on contact centers, where they use voice biometric for authentication reasons, but also on contact centers that humans are doing the authentication. And take for example retail Retailers. Especially during peak periods, they experience elevated risks of fraud.
Speaker 2:It's a time when exhausted agents and overwhelmed systems can become prime targets for fraudsters, since the bustling activity can often mask those nefarious activities and it really puts retailers and customers at risk. With the advent of the technology, by integrating AI-driven authentication methods, one of which is voice biometrics we're not saying that's a single factor of authentication, but it's a strong additional factor of authentication we can really help retailers significantly enhance their defense against such fraud attacks. And these systems are very quick to verify a customer's identity, they can reduce the risk of impersonation, excuse me and they can also detect deep fakes, which is really important, right, and they actually perform much better than a human at detecting a deep fake, and that's really why we are so much advocating that it is mandated in this day and age, with deepfakes being so accessible to everyone, it's mandated that we guard contact centers with these anti-measures against fraud.
Speaker 1:Wow, that's really on the cutting edge here. Do you have any real-world experience yet on using voice biometrics to detect fraud and what's the accuracy you've seen in the real world, which is always a challenge versus the lab or in-house?
Speaker 2:You'll be surprised how much fraud goes through the contact center. It's, it's, uh, it's really really surprising. Um, uh I forget the source of the analysis that I'm going to quote, but around information about an account so that they can eventually do an account takeover and they will speak with different agents at different times and take fragments of information from each, but that every day, not just once or twice or in some odd cases. This happens as we speak in all contact centers, and different contact centers have varied levels of defense against that. Another thing we're doing for our customers, especially in the financial sector, is behavioral analytics, right? So if you see that some account is being under attack, like trying to break into the passwords time after time you know these kinds of things in the old days automated systems would just ignore. They wouldn't think there's something funny happening, right? But with behavioral analytics, along with generative AI and deepfake detection, we're able to put all these factors together and provide a fraud score which can then be acted upon in real time and stop the fraud attempts at their tracks.
Speaker 1:Wow, that's phenomenal. And I can say, with deepfakes it's scary and amazing. I actually have created a deepfake of myself. There's a tool called HeyGen and I use it to create marketing videos and it works, frankly, perfectly. I mean, I sent a couple of the videos to my wife and my mother and they sort of looked at them and 100% thought it was me, with my voice and my mannerisms. That's amazing.
Speaker 1:And this is commercially available. You can train it in 10 minutes and luckily I'm a good guy. But you can just imagine the potential for misuse and fraud with these sort of things. It's going to be shocking. You know, what else can the industry contact center industry and beyond do to tackle fraud with AI? Obviously you have a suite of tools, but what's the big picture? What else can be done here?
Speaker 2:It's an arms race. It's an arms race really. You have to keep up with the technology as this technology becomes more powerful and more accessible to just everyone. I mean the example you just gave Evan says it all right, and I do hope I'm speaking with Evan and not your deep fake.
Speaker 1:This year, next year maybe, maybe not, we don't know.
Speaker 2:But yeah, you just have to keep up with the technology. That's all what enterprises need to do.
Speaker 1:And speaking of keeping up, I mean, how does Amelia keep up? Stay ahead, really. In AI technology and innovation You've managed to be on the cutting edge. How do you keep that edge?
Speaker 2:The hard way we fight fraud and we fight problems every day in the field. Having access to real fraud happening in front of your eyes and among our customers, I think or fraud attempts, I should say that gives you the data you need to build the technology to combat those attacks, and it's ever-evolving. So we have to keep analyzing and keep understanding different behaviors, different technologies, different ways that fraudsters are trying to break in, and just keep up with that and be proactive.
Speaker 1:It's a cat and mouse game for sure. Maybe shift gears. Talk about the team you've built over the years and kind of the culture. Not many people know about Amelia or actually know much about tech in Greece, so tell us about Greece. So tell us about the team and how you're growing.
Speaker 2:Well, so we started out as two people me and my partner Pelias out of, not the garage of my parents' house exactly, but the room above it. So, and that was 22 years ago. But you know, we had a vision and I think I was very lucky in my life to find a lot of good people, of talented people who shared the vision, and they came along for the ride and it's been a great ride At Amelia. The DNA of our people is like we all love a good challenge and we like to apply ourselves to solve real business challenges and that's why we do it. That's what gets us all 360 of us what gets us out of bed every day. And you know, providing value to customers is what it's all about. So we're very customer focused and we drive the technology both from a vision perspective. Both from a vision perspective, but also from the perspective of having our eyes and ears close to the market and understanding what requirements our customers have and making sure that we can provide value around those.
Speaker 1:Well said. So you're also global. I've seen you at events here in the US and throughout Europe. Of course, yeah, you can kind of get to Athens direct from anywhere in the world, which must be helpful for business trips. What are you looking forward to over the next weeks and months beyond summer vacation, which I know is very luxurious and long in Europe, but what else are you looking forward to professionally over the next few months?
Speaker 2:So Greece is where some of us live, but our businesses, amelia's customers, are in North America mostly, and then around 17 different countries around the world. So we're quite an international company, focusing in North America, germany, uk mostly, but, as I said, 17 countries where we're already active in and we support Our people are all over the world these days, right, so we have remote employees all over the world. The only requirement is to have the right attitude, to share the same values as we do and to be able to work in a time zone that suits our customers. That's what it's all about, right. The rest is technology. So I've been very lucky to be able to live in Athens in Greece and have my business outside of Greece, because Greece, of course, is a very small market.
Speaker 2:We're very excited about the drive-through business. This is a relatively new endeavor in our journey as Amelia. We've always been active in the contact center, but this is something which is outside the contact center, and we've been working on this since 2019. We were able to take our foundational technologies of speech to text, text to speech, natural language understanding and and wire them in a way, in a different way, to solve a different kind of of, and it's a very challenging task to take a quick serve restaurant order in the drive-thru. It's a very challenging environment. There's a lot of noise, the way that people order in drive-thrus it just blows my mind. The complexity of ordering is nothing like anything you would see in a contact center, for example. So you know it keeps us on our toes, it keeps us solving new challenges and we are at the stage where we are scaling that in California today and across the nation soon. So that's something really exciting for me lately.
Speaker 1:Well, that's fantastic. Can't wait to see some of the data and technology behind the scenes making that work. We'll save that for a future episode. Well, thanks so much for joining us Really intriguing and insightful discussion, and thanks everyone for watching. Reach out to Amelia. They put out some great content, very educational content on social. Give them a follow, thanks everyone. Thanks, dimitris, thank you, evan. Take care. Bye-bye.