Elad Ferber wants AI agents to answer the call

0:00Starting a business comes with its share of ups and downs, which is why staying true to your vision is essential. a non-negotiable for Romeo and Milka Bregali, Capital One business customers and co-owners of Ra's plant-based restaurant in New York. Romeo and Milka took a leap of faith when starting their own restaurant, gutting an empty space and building it from the ground up. Every pipe, every wall, every detail. But building from scratch came with a heavy financial burden, which is when they turned to their Capital One business card. With the flexibility of the card's no preset spending limit, they were able to spend more and earn more rewards while bringing their vision to life.

0:36Today, Raz's success is proof that with passion and the right support, it's possible to make your dreams a reality. Learn more at CapitalOne.com slash business cards. Pioneers of AI is made possible with support from Inflection AI. It's not just enterprise AI. It's your enterprise AI.

1:00Please hang up or hold for operator assistance. We all know this sound. It's the soundtrack to every phone call you wish you didn't have to make, to the car insurance company, or to that medical specialist you need to send your kit to. Yes, it's really still part of life in 2024. A lot of the default communication channels are voice, are fax, and, you know, it feels antiquated. Actually, that's the world we live in today. Elad Ferber is a computer engineer who's applying the power of AI to one specific area, the healthcare system. He's co-founder and CEO of Synthpop, a company developing AI agents that say, let your doctor's office talk to your insurance company.

1:56Right now, administrators spend hours of their day on the phone verifying basic information. and even faxing documents. The goal of SynthPon is to save medical offices time and money. We actually come to the healthcare system not trying to change it per se, but you know what? We accept you as you are. We're just going to streamline processes with your current modalities and with your current problems.

2:27Today, we're talking with Ilad about his company and specifically about one of the coolest AI agents I've seen to date. We'll dive into how AI agents like this can save valuable time, the ethics of an agentic AI future, and just how this application of the technology works. I'm Rana El-Khalyubi, and this is Pioneers of AI, a podcast taking you behind the scenes of the AI revolution.

3:08Hi, Elad. Thank you for joining us today. Hi, Rana. It's great to be here. Thank you for having me. So this is not the first time we're interacting. In fact, we met about a year ago when you were raising money for Synthpop. Right. And full disclosure, I'm one of the early investors in Synthpop, very proud to be a supporter of the company. One of my investment theses is the intersection of AI and health and wellness, and you squarely sit in that space. So Ilad, give us the elevator pitch for Synthpop. So at Synthpop, we come to automate healthcare workflows and specifically administrative workflows.

3:46We really believe there's so much work that is being done today that can be automated. These are mundane tasks that are really tedious, to be honest. wasting hours of people's days every day. And I founded the company together with Jan Janink, who is a computer science PhD from Stanford and taught scalable systems and applications of large language models for years. And we just decided to attempt to solve this like huge problem in healthcare that drives so much cost and so much frustration, not only for administrators, but also for patients. And that could mean anywhere from patient intake and understanding, for example, who the patient is.

4:27We need to type their name into our system and see if they're there. And if they're not, we need to create a new file and we need to ask the patient some questions, maybe have them fill a questionnaire. And then we need to call insurance or log into a portal and figure out if their insurance is active. If they're coming for a specific service, what's their co-pay, what's their financial responsibility. And that's even before you see them for the first time. Now they come to the clinic. Okay, we need to take you in and we need to figure out what services we want to provide to you and delivering those services.

5:01For example, we support a customer that is building wheelchairs for patients. A power wheelchair can have 10 different vendors that are supplying different pieces of the wheelchair and they're integrating it together. And every piece needs to be justified with Medicare, for example, to be eligible for reimbursement. We do all of those things and all these things you can imagine. They take so much time. And by the way, across this continuum, it's important to use modalities in our mind that humans use. That means vision. That means text, obviously, because that's what most people think about when they think about AI.

5:43Exactly. But also voice, listening and speaking is very important to us. And AI actually has an opportunity to really transform communications through these modalities. And that's what we're here for. One of the things that I love about SynthPop is that you're building AI agents, which I believe is the next frontier of AI. So that's basically AI that isn't just, you know, you're not just chatting with AI. It's actually getting stuff done on your behalf and it's automating these mundane tasks. Yeah, well, we have a bunch of products out. One of the latest products that we've released and is now making many thousands of phone calls to payers every single day is our AI caller.

6:26And essentially there, payers have varying level of support they provide via portals and APIs. And you can imagine a sleep clinic that is covering several states. So they have multiple locations and they can send sleep tests to people's homes and they can also provide therapy. you know, it's located, let's say, in California and they serve multiple states, they need to now navigate many health plans because they serve hundreds of thousands of patients a year. And every patient has their own plan. Every patient has a different copay and deductible. Maybe you have a patient that's in Blue Cross Blue Shield of Nevada.

7:04They're covered there, but actually living in Arizona now. And you need to figure out which plan guidelines are actually active for that patient. A customer like that will have a team of dozens of people that what they do is actually just wait on the line, waiting for 45 minutes. We've seen people waiting for an hour, hour and 30 minutes, just getting validation for patient's insurance. And it's insane. And our bots can do that for them. How's the AI caller going to do that then? So it calls the payer. So of course, we also are able to handle documentation intake, right? I mean, was our first product.

7:39And so we can read the insurance card, we can understand who the patient is, we can understand what services. So our AI also is connected to their system to understand and actually retrieve all the information it needs, which is also kind of a fascinating thing. So if there's a question on the other side, we know where to get to get that data, very much like a human operator would. I think that's really important, right? Like this AI caller, it's not that it has a script that it's following. And if it deviates from the script, It's stuck. The agent can decide what to do. So actually our agent at any given point in a conversation is choosing what the best action for it to do is.

8:18And it could be to go and retrieve information. It could be to dial a number on the dial pad, or it could be to ask a question. The agent can decide in real time and it does a great job doing that. We have a great success rate with those calls even today. It took us actually a long time to get to very high success rates, but now it's doing a really good job. I love it. Well, let's try it out. Yeah, let's do it. Welcome to the Anthem Blue Cross Provider Service Department. If this is a medical emergency, please hang up and dial 911. Your call may be monitored or recorded for quality assurance. What you're hearing right now is a call that Elad recorded, with permission, between an insurance company and Synthpop's AI agent.

9:041-1-4-4-7-6-7-7-7-7-7-7-7-7-7-7-7-7-7. That voice right there is the AI agent. Does it sound familiar? It's actually a clone of Elad's voice. After the AI agent gets through the robophone tree, it's directed to a real human on the other line. Thank you for calling. Anything provided to Ruben says this is, Ben, how can I help you? Verifying whether prior auth is needed. Sure, I can help you with that. Can I have your first name and initial to your last name? Felix R. Philip. All right, thank you. Felix R. The AI agent can provide identifying information and also tell the real human on the other side of any information it doesn't know.

9:57Okay, thank you. And do you have email for feedback purposes? I don't have it. That's okay. No worries. Like Elad said, this process of navigating insurance companies can normally take hours. This AI agent saves precious time. Thank you. Have a great day. You're welcome. Bye for now. Thank you. Bye. This AI agent in action is pretty cool. It's responsive, it's accurate, it's wildly efficient. But how does this technology work? And what guardrails are in place? That's after a short break. Stay with us.

10:59If you've spent any time building AI products or leading technical teams, you know this. Transformation doesn't fail because of ideas. It fails because teams can't move together. Enter Atlassian's Teamwork Collection. It has planning in JIRA, documentation in Confluence, video updates in Loom, and now AI agents in Rovo, which connects the dots across your work so nothing gets lost. It's one AI-powered teamwork platform designed for how modern teams actually build. Learn more at Atlassian.com slash TeamChanger. That's A-T-L-A-S-S-I-A-N dot com slash TeamChanger. I'd love to take us behind the scenes on how you've built this AI agent.

11:51So this AI clone is basically a clone of your voice, right? Yeah, that's right. We needed someone who we had the voice rights to. We didn't want to take Scarlett Johansson or something like that. So my voice was a fine second alternative. I think right now, as we speak, there's thousands of calls going on with my voice, which is pretty cool. Yeah, it is pretty cool. But is it, you know, I guess you have to really put trust in this AI agent if it's going to have access to information, right, about the patient. Where can this go wrong? I think it can go wrong if it misrepresents the patient or if you give it tasks that actually have a more substantial financial impact.

12:33Like in this case, we're just finding out information that should be available to our client and should be available to the patient. Like is prior authorization needed for a specific therapy under the patient's plan? Like just asking that question doesn't change anything for the patient, but we do QA on these things. QA, as in quality assurance. So we have a QA team that actually listens to a lot of these calls. We do it also automatically using AI to QA our AI. And so we make sure that the accuracy level on those is very, very high. We do not tolerate types of errors that will misrepresent data to the patient, or we'll say that something is covered when it's actually not covered, for example.

13:20that's a big, like we actually prioritize and we can tweak our algorithms to make sure that those mistakes are almost non-existent and perhaps other things are maybe more tolerable. Elad says that the AI agent only uses information that is already in the system. So it's not prone to any errors that are out there, say on the general internet. But mistakes could still happen. Like the AI agent could give the wrong code or the wrong state because the information is wrong in the system. Maybe there's an old insurance card on file or an address that hasn't been updated. These are the same mistakes a human agent would make if it was the only information they had.

14:06We don't do a job that is better than a human in that sense. There's some mistakes that are just inherent in the system and that's something that we are not necessarily able to fix with this. Walk us through the behind the scenes of the technology. How did you build this? So I think modeling a conversation, you could think of a very simple loop of trying to figure out when the other end has stopped speaking. And that's something that's called silence detection. And when you talk to chat GPT on your phone, if you've tried, one of the key things is to understand when it's time for me to speak. And actually, even for humans, it's pretty complex because we can talk over each other.

14:49We can understand cues in our tone to understand when to talk. And for it to not sound uncanny, that's something that we pay a lot of attention to. I mean, it was interesting. It was saying, it was adding all these things that makes it sound very human-like. Exactly, because latency is everything. And we as humans, we're so similar to LLMs in a sense, because when we decide to come and speak, when it's my turn in the conversation to speak, it's not like I have my entire soliloquy that I'm going to speak written out in my head at that point. I'm actually also almost like a word by word generator, kind of like an LLM, right?

15:27And so this um, and this like giving the model a little more time for it not to think, not to sound uncanny and to have like the snappy latency that the other side knows that we now it's time for us to assert and to speak in a conversation. That's really important. And I mentioned earlier in our conversation how we have a model that the agent has to decide what action it needs to take at any given point. So should I speak? Should I retrieve? Should I hang up? Also, maybe is the conversation done? There's a bunch of actions. And I think understanding the right action, that's the number one key thing.

16:05There's so many nuances. We had for the longest time on Blue Cross Blue Shield of California, we were saying and spelling patient details. And it would not get it right on the other end. You know, E-L-A-D, a lot. Like, that's the name of it. It was like A-L-A-D? No, it's E-L-A-D. And I would totally hear it and understand it. And then, for example, we started doing NATO Alphabet. and it fixed it. Like E for elephant, L for Lima. Yeah, exactly, exactly. And some of the small tweaks that perhaps are non-existent in other use cases, but we found them and that unlocked Blue Cross Blue Shield of California.

16:44You know, thousands of calls are all of a sudden the success rates goes from like 60 % to 95%. Like there is an agentic flow that you need to devise. And I think building agents is actually not simple. It's custom work. Wait, explain to us what you mean by an agentic flow. It's like an algorithm that can contain, in our case, 70 calls to LLMs along the way in order to do one big task. But it has five big pieces of the flow, and that's maybe fixed almost in code or in configuration in our case. But within each piece of the process, there is some freedom of operation for the decision-making process to do more or less, for example.

17:28So when we say an agent, we sometimes think of an LLM that can autonomously decide what to do. We partially have that, but actually also it's partially scripted. Which is important in this case because it's a very set kind of sequence of administrative tasks. Yeah, and you want to have also repeatable results. If you give AutoGPT, create a business that will make me a billion dollars a year, it will come like with something else. If you just tweak it with one word, it will come up with something completely different. And even if you write the same prompt, it will have completely non-deterministic results.

18:04For us, we want to guarantee less of that. And we want to be more deterministic rather than not. And I think that helps us achieve that because we want to make sure that we can deliver something that customers can count on. We call it composable flows. And I think being able to have composable flows for custom use cases is really important. By the way, we have a composable flow that we can actually adapt our agent to do what your humans are doing. And so our customers can start with our AI agents in a small scale. It works alongside humans. They don't have to give us 100 % of the volume they want.

18:38Our agents actually work across the same process, the same instruction set as a human would and achieve the same format of results. You know, humans and AI work side by side. And what we found is as soon as we can show that, the customer is like, we need to go full throttle right now. I'm sure some of our listeners are listening to this and thinking, here we go again. Here's another example of AI taking human jobs. Like, what would your answer to that be? Well, I think that there is going to be an impact of some sort. But to be honest, the people we work with, we still need them to a large extent.

19:19And we work in tandem with them. And the call scenario is an interesting example because there are some complex scenarios of primary, secondary, and tertiary insurance problems that we might not be able to solve. One of the actions our agents can do is say, hey, I think this is a little bit above my pay grade. Let me loop in an expert. And I think that is very powerful. We still need humans to work alongside us. We need humans to train us. I think where there's going to be some reskilling in the workforce is actually in those more entry-level jobs, not the experts and those who spend like the last five or 10 years doing administrative roles, because the expertise they built is actually super valuable.

20:04And it's actually that workforce that is so hard to staff, that is relatively low entry level and is highly cyclical. You know, there's a lot of turnover. A lot of our customers really struggle with that. And I think reskilling is just a natural thing when you, you know, invention of the car. You don't need as many horse caretakers anymore. And when they move to the electronic health record system, like maybe you can do away with a lot of mail and shipping actual stacks of documents between offices. Like those jobs are transformed. But again, I think the healthcare experts, I don't think they're at risk of this.

20:44And actually, they can just be much more efficient, less burned out. And I think that's what we see. And the reactions are all positive from people we talk with. So another question I have and I'm curious about is, does Elad AI disclose that it's an AI when it calls, you know, an insurance agent or it doesn't? Yeah, so that depends. We had a kind of an interesting experience where I think that if you're talking to a human, there is an interest in disclosing that, right? If I'm talking to another bot, I think that's fair game. I'm also kind of fascinated by new business models that come with AI.

21:25So what is the business model in this case? Is it per call? Yeah, we thought about it so hard. The most important thing for me is ROI for the customer. We don't charge per minute or anything like that. We charge per successful call. And what we call successful call is a human would not need to do that. We have just eliminated the need for a human to touch this ever again. Job done. Yes, that's true also on our order pipelines and our document ingestion and data entry. Sometimes we will need a human in the loop. And sometimes the order is so complex or unreadable that, you know, we can't do anything with it.

22:07We will basically align our ROI with the impact on the customer's labor. we wanted to make sure we're priced in a way that makes it a no-brainer for the customer to use us if we were to take like a different approach of like a big chunky platform fee and paper minutes for everything we do for you i feel like customers are going to have a harder time committing to that because you're delaying the roi discussion the roi discussion is going to be there roi leads the way And that's what we believe in. So that's how we kind of decide on our business model. We're going to take a short break. But when we come back, Elad and I talk about what an agentic future could look like.

23:13Meet Nicole Nicholas, Capital One business customer and co-owner of Ansett Uncles, a plant-based restaurant and community space in Brooklyn, New York, that got its start from a need for unity. The inspiration, it was born from the desire to create a space that felt like home, where we can connect community culture, good food, and come together with family and friends. That's how we birthed aunts and uncles. Nicole and her husband, Mike, were fulfilling their dream of bringing people together out of their home kitchen, but they soon learned that the demand for community was greater than they knew.

23:44It became overwhelming and we were like, we need home, but not in our actual home. We realized that there was also a need in our community for something bigger and in our neighborhood, so we had to find a place. Moving from a home operation into a storefront was a huge next step, but Nicole and Mike were able to take it on with the help of Capital One Business. It's not for the weak. As a small business, finding resources is super important because that's the way you'll be able to manage and scale. We would have never done that without having Capital One to be able to help us along the way. The cashback rewards are very helpful.

24:20You know, it just gave us that runway to be able to breathe a little bit. Then you get to focus on the cooking of the food and making the experience great. To learn more, go to CapitalOne.com slash business cards. So SynthPop is really focused on building AI agents that automate these kind of mundane tasks in the healthcare system. I want to move us on to what is the place for AI agents in our everyday lives? I'm curious where you think the opportunity for AI agents is for consumers, and where do we stand with that? Have you seen any examples that have impressed you or you found curious? Yeah.

24:59So I think I'd like an AI that can handle my communications, but I actually would like it to ask me maybe five questions that will help it actually understand my mood today. Because you know what? Maybe I've had a crazy dream. I have a new vision about who I am and what I want to do in this world. And actually, I'm going to answer those emails a little differently than what I've done so far. And that's part of the friction today with adopting AI tools that can scour your emails and try to kind of get your style. And, you know, I've tried Gemini AI, even chat GPT, I try to kind of draft an email.

25:33I rarely just accept what it is and send, not because I have a problem with AI generated text necessarily, but because it actually doesn't exactly capture my emotion, my vision, my purpose. And I feel like being better at that requires some back and forth with who you are at the moment because we are living and breathing animals and we're changing from moment to moment. We're doing this interview in the morning. If we're doing it in the afternoon, it could have been a little different. And so I wanted to capture me as I changed throughout the day. And I think that sensitivity to the other side is critical.

26:12Yeah. As a CEO, what's keeping you up at night? I can sleep pretty well. I think after having two young kids, but what keeps me up at night, maybe metaphorically, is how fast to expand. That's really kind of interesting problem because always kind of keeping that tension of being able to serve and expand with our current customers and learn to do that better versus rapidly expanding to the blue ocean of opportunity and getting more and more new customers and new use cases. And yes, we've built a composable and scalable architecture, but it still takes time and effort and technical work to serve those markets and new customers, that it's always like an interesting dilemma.

27:02It's a blue ocean out there and we need to go after it and we're going after it. But what is the best way for us to do that? That's kind of the interesting pull. Thank you a lot for joining us. This was awesome. I learned so much. Thank you so much, Rana, for having me. This was great. There is still so much to develop when it comes to AI agents. Synthpop is making AI agents for businesses. But how cool would it be to have a consumer-facing AI agent to schedule, say, your doctor's appointments? I still need to schedule that colonoscopy. On this podcast, we're going to continue talking about what our agentic future will look like.