Village_Global_Podcast_Why_AI_Agents_Can’t_Be_Trusted_Yet_And_How

0:00I was telling her that like, you need to use the system. It's just better for you, but she had no reason to believe that it is better. She had huge responsibilities in her hand and she needed to deliver that responsibility and she felt that deeply. And all I was telling her is that this is better, but I have no proven point. She didn't not trust the system. She didn't, she didn't trust an unproven system in order for people to adopt the systems that we design as engineers. They need to trust it. I use my credit card and I tap on it and I walk away. Nobody is concerned.

0:30It just happens, right? Like, my credit card could be Canadian, and I could be in San Francisco, or I could be in Spain, and I would do the same tap, and I would walk away. And nobody would trace me, nobody would say any single word. Why? Because everybody trusts that the money is going to go to the right place. We need to do the same thing for agents. When I tell my agent, do something, it's going to do it. And if it doesn't know how to do it, or if it's unsure, I know that it's going to tell me.

1:05I'm Ann Duane, and I'm here today to welcome Moe Katib of One. That's withone.ai. And Moe's been working for 20 years on software integrations. But recently, over the past year or so, he's been agent-pilled. And now, at One, he's doing the integration work that we all don't want to do. And he's super-powering agents and people. So we're going to talk about his decision to open source all 54,000 integrations that he had assembled. And then we're also going to talk about a really optimistic future of work. His vision is that we won't be chained to our computer and that we'll have agents that we can trust. Welcome, Mo. Well, I'd love to hear a little bit more about a few of the stories that have shaped your life as a person and a founder. Yes, absolutely. So I am originally Syrian.

2:01born in Damascus and really what shaped my story. There are about three stories that I can give you that shaped my story and what made me who I am and what made me want to do what I want to do and what I do. So, believe it or not, I've always been entrepreneurial and I didn't actually know it. And so, the first story that really shaped my...

2:26my life was when I was a kid, back in Syria in Damascus, my family, my father is a businessman, my grandfather is a businessman, my uncles, every single one, they're entrepreneurs, business people. So I grew up around that sort of environment. We didn't have the choice where in the summertime, you're either going to school or you work. So you have to pick one or the other. And in many cases, I actually picked both. I would study and work.

2:55So early on, you're brought into that environment even when you were a child. So I would go with my father to his job, and you're engaging with a lot of these business people. But at one point in my life, my dad thought that it would be good to give me a job that is really harsh. Because he's like, OK, if he works at a harsh environment, he's going to become a better man. He'll be able to handle things more effectively. So my uncle owned a chicken factory.

3:24Like, this is everything from literally raising the chicken hatchery all the way to slaughtery house, the whole mechanism. The chicken value chain. The whole chicken value chain, exactly. So anyway, my dad goes to, that's my dad's uncle. He goes to him and he goes like, OK, so Moe is looking for a job he wants to work. And then he tells him behind the scenes, I knew this later, that they basically Scripted a job. That's the most difficult job that not even a single person would actually normally get it So they put me in every difficult place like literally I'm like, so anyway, I'm working in this and it sucks. It's really terrible You know like you stink and all this but you learn quite a lot of different things and as a process What ends up happening is I'm working with the chicken so I have access to really high quality chicken So I bring home some of these chicken my dad tells my

4:17Ants and then ants tells their neighbor and before I know it I'm like transferring 100 kilos of chicken on a daily basis And I'm like this little kid and I'm carrying these like weights of chicken like this sucks But you're not gonna tell my aunt I'm sorry and I can't deliver it for you because you know like you can't do that And then one day you know like I delivered for my aunt the chicken and it is like really big chicken and So my aunt is cutting them and I see she's doing a lot of work cutting them off and I'm like I'm like and you know like what if I What if I get you this already sorted out, cut in pieces, and this all covered and packaged in a very nice way? So you take it, you put it in the freezer, and off you go. Would you pay me like 10 Syrian pounds per kilo? I'm just like, hell yeah, 100% I would. And then that's where my entrepreneurial mind started hitting. I'm like, wait a second, I can make a lot of money from this. So back at the factory, they do this, but at large scale. So they would do tons on a daily basis. So if I threw 100 kilos,

5:14It just goes through, they don't even notice. So I started doing this. And then that became more convenient. And then my aunt started telling her neighbor, that was not a thing, by the way, at the time. You can only buy whole chicken at the time in Syria. Innovation. Innovation, yeah, exactly. Because that manufacturing was for restaurants. And actually it goes into the production lines. It's not to be sold for individuals. It wasn't a thing. At least in Syria, it wasn't a thing.

5:42And then, so my aunt tells her name. My aunt is one of the... I love her to death. You know, she's amazing. But she has a lot of big circles. So word of mouth. She started telling everybody. And then everybody started... Before you know it, I'm like selling. And I swear to God, I thought that I was stealing money. Because I was making so much money and I'm like, I don't know what to do with it. And I'm like, I can't show it to my dad. So that kind of like was one story that like...

6:10kind of made me understand that there's a problem if I had a solution for people would be willing to pay money for it. And it was a good lesson at an early stage. But really, the thing that really got me to where I am today is my love for anything technology. And I love science so much and technology. I love to understand how things work. So early on when I was a child, I got in love with electronics. So I started, you know, tinkering with electronics, like building my own circuits and like solving different problems.

6:40And one thing that really shaped the reason why I immigrated and I came to Canada was because of the story. It's like the rumbling or the starting point of that. So back in Damascus, Syria was basically communist at the time. It was pretty bad. Of course, I knew a little.

7:06Because there are certain things you're not aware of. A chicken capitalist. Yeah, I was a chicken capitalist, exactly. And my family, they were all capitalists. So my family was not the family that was welcome at the time. We were actually pretty persecuted even after. During that time, you didn't have running water. So you have to essentially, the government will get you water at a certain time. And sometime you have many days that you don't get water. And then you had to pump the water.

7:35from the bottom of the building to the tanks so that you could have a reserve. And that process was a pain because you essentially have to first figure out if there is water. And they give you a schedule on the TV. They say, like, oh, this area, this area, there's going to be, like, at this time, there's going to be water. But they don't always. It doesn't always. It's not always the case. They'll say, like, it's 10 o'clock. But it actually arrives at 10.30. So you have to check if that there's water. Because if there isn't and you turn on the pump, it'll burn.

8:06And then you have to also make sure that the tank is empty. Because if the tank is full, you turn on the pump, there's nothing to go, it'll blow up. It's really bad. So usually people start shouting to each other. There's water, there's no water, turn it off, you know, and they're shouting. And this is like very common in the area. You start hearing it. There's so much noise, you know, like. So I thought there has to be a better way that makes this a lot better. So I created a device, two devices actually, that communicate with each other.

8:36So if there is a water you install the first one on a pump you install the other one on a tank and it just works It's like no need for human communication. It just works so my grandfather was a biggest believer in my inventions and He was like do it Make it work, you know like put it on and like he believed in me You know like and of course, you know, it's a big risk because if we could you go out of water if it doesn't work so I put it on and it works and it solved that problem. So my grandfather then tells the neighbor, neighbor started asking for orders, I started doing it myself. And it was all me, you know, like I built the whole thing, like I did the circuit, I drew it down like and solved it. And I used to also put signatures inside my designs, certain things that means nothing, but I know it's my design.

9:25So sold a couple started getting like a bunch of orders and I would go and I put them on and like It felt amazing. It felt really amazing. It's not you're making money But it's not the money that made me feel really it was the fact that you actually solved the problem and it made you made the life of people better So one day I'm in in electronic market. We had an electronic market in Damascus I go there and I want to buy my my like little devices and then I see a really good-looking device that looks exactly like my idea. So I panic. Like any founder, you know, you're like, this is what I made. And then I buy it, I go home, I open it, it's my design. And I know it's my design because they added the thing that doesn't do anything. And I'm like...

10:15That sucks. So I actually remember I went to my dad. I was crying because I brought it to my dad. My dad doesn't know what I'm showing. I'm like, look, this is my design. And he was like, what do you mean? I'm like, they stole it. They stole it. I want to sue them. And he was like, where do you think yourself? You think you're in the US? We can't sue them. So the company turns out to be one of the close to the government companies. They essentially mass produced it. And so my dad was like, just make another thing. And I'm like.

10:44This sucks. This sucks. It's like always the same. And I started noticing the injustice. And I started coming up with, I'm like, I can't live here. I need to go. I need to leave. And US and Canada was my top pick. And then I migrated to Canada. OK. And it sounds like in Canada, you started tinkering with software.

11:08And so what was your first software product? And then maybe how did that arc into what you're doing today? I started working in Canada for a school, great school. And I was the head of like lead engineer and I was tasked to build the software that manages the school from the ground up. This is everything to do from students coming into the school all the way to graduation and then everything in between. This is like managing the staff, managing payroll, managing invoices. And it took me a long time to complete the whole thing. And this is at the time, by the way, where we had our own racks. We had our own server. I built my own network. Excel was a thing. And Access was a thing. It was around that time. You shared files. And if you had two people opening the same file, they corrupt. So I started building that software. And really, there's a story there that

12:05taught me some a lesson that I never forget and I still use that lesson in today in everything I do today and We used to have this Really hard-working single mother. She used to be In finance takes care of like collecting money Making sure that the finances. It's like CFO and she would work seven days a week like that never takes a time off You know, she's working like 10 sometime 12 hours a day on and on and on and like super dedicated woman you know like I have a huge respect for her and we built this software that's like so much better and I knew it's so much better because like it's not because I designed it I know factually I'm an engineer you know like I know it's better and I found in the old systems many leakage and of course you know like you have a system that's excel and access and like it's gonna you're gonna miss a lot of things so this system was complete in the sense of a SaaS

13:06And I told her, I'm like, okay, we have the system. I trained her on the system, and just she wouldn't use it. So I do more training, I do more training, and she keep using the old system. So on the old system, whenever she uses the old system, I have to re-sync with the new system, re-migrate. It's a lot of work to do that. Sometimes a week of work, I have to do this. And then when I'm done, I'm like, okay, you have to now use the new system. I'm not kidding. And then she would always come back. So one day I did the unspokeable. I basically cut off the old system.

13:35Entirely and I swear she she I'm in my office and she comes screaming She's like I can't access this because I put up I put a message contact more and She came to my office went to the CEO. She was really upset CEO comes to me is like really you cut off the system like I cut off the system I I told him I deleted it, but I actually didn't need it, you know, but and So She was like, OK, well, you have to work with her. You have to get her comfortable. So I started doing this. And in time, she started using it. And then slowly, she started working like six days a week, then five days a week, and then eight hours. And then in two months, she took her first vacation. And she came to me thanking me for that, because for two years, she wasn't able to take a vacation. So I failed her because I was telling her that you need to use the system.

14:31And it's just better for you. But she had no reason to believe that it is better. She had huge responsibilities in her hand. And she needed to deliver that responsibility. And she felt that deeply. And all I was telling her is that this is better. But I have no proven point. So she didn't, like, she didn't not trust the system. She didn't, she didn't trust an unproven system. And that's really important. That's like a distinction that I take with me is like, in order for people to adopt the systems that we design as engineers.

15:00they need to trust it. And you can't just tell them, trust me, it's good. It needs to be a process. There's a lot of education that goes into it. There's a lot of talking. Like, I didn't sit down with her to see why, you know, like, I mean, I was just telling her, you should use it. It's better for you. It's like so much to look here. I'm showing you here. But I didn't sit down to see, like, where is her anxiety? Where are the points where she felt anxious? Because there were true anxieties in her, you know, like when she came screaming.

15:26And she's not a lady that screams. And she was screaming. She was literally screaming. And it taught me a lesson that you have to value the people that we serve. They have deep emotions to the way they do things. And if we don't handle these emotions and we ignore them, we will not be able to build better products. Because at the end of the day, the product we build, our product design for human.

15:54Yeah, and human are complicated if their emotions they okay, so that sounds like it might inform your vision of the future of work And I'm curious what that taught you and how that impacts how you build with agents today I don't think this is this goes as a surprise work is gonna change and you know the world are Very noisy right now, and I don't like the noise because I feel You hear a lot about, like, I 20x this, I 30x this, I 50x that. I am like, now I'm 20. Let me show you my recipe that's going to 500 whatever x. There's always some x to something. And this acceleration of maximizing whatever we're doing. And OK, this is great. I'm not against that. But what I'm against is this sort of hype that is created and how it's impacting the people who don't actually understand the verbage.

16:55what they feel, they feel that they're left out. And they feel that there's not much they can do. They're like, I don't know what's happening here. And it feels like you are on a horse and you're seeing the car, but you cannot get on the car. And you're like, I can see the car. But the problem is, the car, people are saying it's a Ferrari, but it's not yet a Ferrari. So to me, the way I think about work is I always come down to the human. And my...

17:23My vision and the reason why I do what I do, as difficult as it is, is because I really want to serve other human. And I truly mean it. I really want to make human productive. It's the reason why I do what I do. Because when you're productive, you'll be able to do a lot more things that you enjoy. And the way that we have defined work, we defined the work in the way it is. You're sitting behind a computer, in most cases. You're sitting behind a computer.

17:52and you're punching keyboards and you're moving pixels. That was our definition of work. And that was great because before this, we used to be on papers. I remember that time. That wasn't very good also. But really, if we come down to it, what do we human want? What do we want? Do we want to be sitting on the computer? It's a big question. It's a very big question.

18:19But if I if I'm thinking about it from first principle like what do human want, you know like they want to They want to have conversations. They want to have they want to they want the work done They they they they want this stuff that they that needs done done. They don't care if it's 20 X or 50 X or 100 X it just needs to be done and Usually the work that needs to be done. It needs to be done in a certain time yet you if you could get a faster will be better There's no doubt about that, but usually like you're not If you need to send an email, you could send an email by six. If you send it five minutes to six, or six doesn't matter, as far as you send it by six, it should be fine. And most people think that way. We tend to be in the valley. There's this sort of feeling that you need to go faster. But I think people want to be away from the computer. And in order for people to be away from the computer, they need to trust the systems that we're building. And the agents,

19:15have that capability to get us away from the computer. Yeah, I think this is a big idea, right? It used to be that you had to be at home or at work to do a phone call, and the mobile revolution changed that. And with agents, will we think it's cute that people used to sit at a desk with a keyboard, a QWERTY keyboard, and give language in that way? Or worse yet...

19:43coded in some weird coding language. He went from designing software for a school system to working on agents today. So describe that arc. Yes, that's a big shift. So my journey really shaped itself in the way that I am today. Unfortunately, because of the war that happened in Syria, my immigration status got... There were some challenges because of the... I wasn't able to get a passport, which...

20:11kind of made my immigration status become very complicated. When all that stuff was sorted, and I mean, you know, like there were a federal judge in Canada that actually ruled in my favor, which I'm so grateful for. I'm gonna send her a thank you letter in time because she changed my life. And I became Canadian. And that was the time when I started my own business.

20:41moved away from the school and started my own business. And in my business, because of how technical I am, I would go more towards problems that are large in their scale. So a lot of the problems that we were solving were migrating from one system to another system.

20:57or working with ERPs and stuff that relates to integrations in general, but more so for the enterprise. So I saw all the difficulties of integrating systems together, and we would work on one integration, literally one integration that would take us eight months and sometimes even 12 months, and we'd have eight engineers working on one problem. And it'll cost millions of dollars to actually get it to work.

21:24So throughout time, I started noticing that there are similarities across different platforms and that you could come up with some sort of abstraction layer that would make the integration problem completely disappear. So that's what I've started working on. So like for the last 12 years, I've been working on the integration problem in that scale. Now, how does that translate from what we had?

21:50from what we were doing, which is pretty much SAS. You think of it like low-code, unification of data. The typical integration SAS that you see, like the iPass, I think they call them. In 2024, we've done a lot of work with AI at the early stage. So I remember even having a very large scale. I would call it like a mind system, where it has little pieces of Agentec.

22:20because at the time you can't trust a lot of things, but you could trust a small thing. Like label this, it would be able to label it. So we built an entire, you would call it Agentec today, but it's more so Agentec workflow that allows us to map data of integrations. And this is at the time of integration OS. Because of that, we generated so much data, the data which is the prompts, like the prompt that we will give to the agent so that it creates the schema that maps one system to another system. And in 2024, I got an idea. I'm like, what would happen if I take this data that I have for Gmail, for example, and give it to an agent? Would the agent be able to send an email on my behalf? This was early. Not a lot of people were doing this. And to my surprise, it worked. The agent was able to send an email. And then I'm like, OK, let me see if it could be able to read from a CRM. So I took another piece of knowledge that I have, gave it to the AI. It was able to read the CRM. So I gave it an execute power and the knowledge.

23:16I became so convinced that this is going to be the new direction because you would start to hear people talking about a genetic agent and like the definition where it was a little fuzzy at the time. Not a lot of people understood what that means. Today is pretty clear. Well, how do you I'm curious. How do you define a genetic software today? To me, a genetic software is an agent is an entity that reflects a group of work that we human do today.

23:45So, and the way I would define it is like an effective agent is like a salesperson, for example, or VA. And then you can add more element to it. But it's like, I have a unit of work that needs to be done, and it needs to be done autonomously. And that unit of work that's being done autonomously will have many pieces of it that must be deterministic. And I'm arguing that that needs to be the case in order for adoption of agentic to actually pick up because it ties into trust.

24:13And I'll dig into this. So to me, the agent adoption requires human trust, which goes back into this whole concept. In order for us to deploy the agents that are actually going to do work, the human need to trust it. Right. To accomplish the goal. To accomplish the goal. And to be able to accomplish the goal. Correct. Yeah. Because we are like in San Francisco, people are willing to take risk.

24:44We adopt technology faster and we tend to be on the cutting edge of technology and we try things that might not work. The tolerance of failure is pretty high. There's no issues there. But most people are not like this. They are concerned. What would happen if it leaked information? What would happen if it sent the wrong email? I don't believe that we have yet the mechanism to...

25:13Like, we don't have yet that sort of system that creates the trust for people to allow a higher adoption of agentech. Because I truly believe a working agent is going to free a lot of our time. It's going to get us into the sort of future that I really wish that it's going to be true. Which is the future where we as human, we go back to what we like to do. We go back to...

25:37talking with other people, going to the communities, getting away from the computer and trusting that the systems are actually working. And I like to use an example. If I go today into a store and I buy a drink, I use my credit card and I tap on it and I walk away. Nobody is concerned. It just happens, right? My credit card could be Canadian and I could be in San Francisco or I could be in Spain and I would do the same tap.

26:04and I would walk away and nobody would trace me, nobody would say any single word. Why? Because everybody trusts that the money is going to go to the right place. We need to do the same thing for agents. We need to make it where when I tell my agent, do something, it's going to do it. And if it doesn't know how to do it, or if it's unsure, I know that it's going to tell me.

26:24And that's really the balance. You don't need the agent to notify you on everything because then you get overwhelmed. Now you're creating more work for me. If the agent is going to confirm everything with me, there's more work for me. There needs to be a balance. And what I found works really well, and this is something I'm building for myself, is you start with a lot of notification, and then the agents start understanding my preference. And I start building up the trust with the agent that is like, now, OK, it's replying for my emails. And at the beginning, I'm like, Under no circumstances, you reply. You show me everything. And then I start saying, yeah, this is good. This is good. This is good. This is good. And now there are certain aspects of things that are completely autonomous, like refund requests, for example. I don't even need to know about it. It just happens. So you have trained it bespoke for you. Correct. And so you have onboarded and trained your agent. Yeah. OK. Well, let's talk a little bit about your company.

27:19It's with1.ai. One came as a result of all the knowledge and all the things we've done in the past. As I mentioned, we had all this knowledge that we accumulated about these integrations. It's not a secret that integration alone, it's not going to be a moat anymore. It's not a moat. You cannot have a company that is just doing integration.

27:45call it a moat or even a sass. I believe software is going to become commoditized as a whole. So I started thinking, where is the value? What am I doing? Where is the value that if someone is using me, what's the value? And I start reasoning over and over that the value is in the capabilities of this. I'm selling the capabilities. And then if I think even deeper, what does that mean? Well, it comes down to effectiveness of these agents. So first of all, I needed to make sure that neither me nor my team are living in a bubble. So I took our what we thought is so valuable, which is our knowledge. And you know, I know I am very biased, but our knowledge is really clean. And like if you use it with any agent tech, it works. That's because we vet the knowledge and like, we have system to scrape the documentation, then we put it through a

28:43So we scrape like let's say for example We want to work with Gmail we go and we scrape all the docs and then we take whatever we scrape We put it into other system subsystem to make sure that the knowledge we scraped can actually be executed by an okay got it this seems like your ideally integrations would be an easy thing for agents to do because there is good documentation in theory but in practice it isn't always the case so you have made sure that it's agent ready and then also agent always ready, meaning you're maintaining it over time. That's right. And I would argue that, yes, the agent can go on the docs and do it, but it's the Wild West because the agent can read the right doc, but it could also read the wrong doc. It could read an old doc. So you are at the mercy of the index. So what we're doing is we're like, no, we actually are vetting this knowledge to make sure that it is actually accurate. And then we run different use cases.

29:39So we run, we have a system called the scenario generator and it generates different scenarios for every action that we use. So for example, I like to use the send email example. If we start from an agent that is like non-train and you tell it like the user could say something like write me an email for Emily. That's what the user would say. And the agent would interpret this and it will say it might write you an email because it might think it's a sample.

30:07and then it'll write the email. The agents now are a little more sophisticated. They'll ask more questions, you know, like, but you could have an agent that is not as sophisticated and it could literally write you an email and the email address will be emily at sample.com. But if that agent has the capability to also send, well, it's sent that email. So you have a tiny little issue there. So what we do is we generate these different scenarios and it goes from like...

30:32Like like someone who is not it doesn't they don't know how systems work all the way to someone who is very capable And they understand exactly like how they're using these tools and we generate everything in between and then we generate the success route and the failure route and then from there we start like seeing the behavior of the the agent like For example, send an email to Emily, well, which Emily? Like, how do I know what Emily you're talking about? Is this an Emily that we've been working on and I have it in memory? Or is this something I don't have and I don't know and I need to look up in the CRM? If I looked up in the CRM and there's like several Emily's, maybe it could figure out from the context, but should it make a guess? And that's the thing, you know, like, is that there's a balance. So for us, what we're doing right now is we're solving the whole schema problem and like, okay, the obvious stuff.

31:19But the scenario generator, what it's doing, it's solving a little bit of a bigger problem. Because as you could imagine, this is going to generate a lot of knowledge, a lot of data. But there's almost always one path that is correct that the user wants. And figuring out that path is not an easiest thing. But if you don't figure that, you will not have high adoption. We know how to do this for ourself. But when you need to deploy this to the masses, it needs to just work.

31:45They don't need to think about like how this works. So like we are building the systems that kind of make it work. Got it. And who is the ideal user for one? There are three ideal users for one that feed into each other. We have big companies. This is like companies who have already massive agent take and massive, in a sense, agent solution such as vibe coding platforms, platforms that are deploying agents.

32:14because they want these agents to be connected to integrations and apps and all this. So we help them. It just works because we solve authentication, we solve the knowledge. And now, as I mentioned, we're getting into the skills. This is where the value really lies. And these are vetted skills.

32:31And then we have the second one, which is San Francisco startup building agents to solve very particular niche. So like you're like sales agents, marketing agents, you know, I would love to work with these types of startups and allow them to kind of figure out, you know, like, how do you solve this bigger picture? You know, like this is like, okay, one agent that's doing sales and like focusing on sales. They, of course, you know, like, they're great ICP for us because their user needs connection and we help them get there. Like we help.

33:00we make it really easy for their user to connect to these third-party integrations. And then finally, prosumers. They use our CLI, and like with a single CLI, you can connect to any tool that you have and build your own agent. You know, like this is the hobbyist, the developers, you know, like people who want to tinker and play with the agents and try it for themselves without necessarily a lot of opinions. And how many integrations do you offer today?

33:29I believe today I think we were at 330. Amazing progress. You had shared previously that the rate of adding new integrations was a certain pace, and then over the past couple months that has really accelerated. Can you talk a little bit more about that? Definitely. The important thing for us is more than the quantity, the quality.

33:55I could easily like right now for example go and add a thousand integrations but the quality would not be as good. We have figured out the processes that are necessary in order for us to actually say that this integration is good. So we start by scraping, but it's not a normal scraping where, for example, when you do a generic scraping, you will miss so much data. For us, we analyze the type of technology that the docs are living in. And then based on these different technologies, we write different scripts that essentially would open different tabs, have different pop-ups. We solve problems such as if you have, sometimes you have a schema, and you have a nested schema, but sometimes you have a schema of a schema.

34:36Meaning you have an invoice that has a bill inside of it, but the bill has an invoice. And if you keep opening, it keeps opening. So like if you tell an agent to do that, it'll basically go in an infinite loop. So we have encountered all these different problems, and we've solved all these problems to figure out how to get to the actual value of the knowledge that we have there and get everything. But that's not enough, because even if you get everything, there would be so much noise.

35:00So then we clean it up. We then have different AI system that actually cleans up the docs so that it's repeatable because we figured out a pattern that if you give to the agent it works every time. Right, okay, got it. And then so you're able to abstract away all the complexity, all the maintenance over time for the agent and ultimately the end user. That's right, yeah. And then, but what we figured out is the balance of the human in the loop.

35:24And that's really key in my opinion. And we see the same pattern happen. At the beginning, literally everything was vetted. And I literally mean it. The first 100 integrations we've added, it was vetted by hand, every one of them. The system were not as good. The models were not as good. And I'm talking this is like in 24. But then we started noticing there's a lot of improvement in the models. And now we're getting into a point where we're getting roughly a 98% success at first run. That doesn't mean, of course, that we handle every edge case. But because it's a network and because the knowledge is open source, you start having contribution. So when, for example, someone could be in Spain and they're using our tech and they encounter an edge case that we've never seen before, we've never even thought about, if they want, they can share that problem with us. And then because most likely their agent would have figured it out.

36:20And because that section of the industry, the ICP that is using the CLI, they tend to be technical, so they tend to correct the agent, which is going to help the other buckets. Because the other buckets, there would be no that human in the loop. So you want to get the human in the loop, which is what we're doing. And ideally, we're getting all the people who use it also become the human in the loop, which feeds the knowledge, which improves the AI. And then we want the AI labs to train on this data. I want them to train on the data because I truly believe that the better the cleaner the data, the better off all of us. And that's the thing. It's a scary thing to open source the knowledge because in my whole team, I had to speak with my team for hours, hours until because...

37:07Our company you know like there's I don't it's not there's no dictator, you know like oh I say this you know I have to convince my team yeah, they built it right like I mean my team built this you know so yeah, so walk us through that decision to open source so I started thinking about the value and I wanted to make sure that I don't fool myself and I found that It's really easy for me to fool myself and into thinking that I have something when I don't have something And so this becomes very important for me as a founder. We as founder, we live in a parallel universe. And we have to.

37:45We have to because we create this sort of vision, and we need to believe it. And if we don't believe that vision, nobody will follow us. Why do people like to work for founders? Because you have this big vision, and you want to make it happen. And you stand for something. But if you're talking with normal people who are not aware, they will think you are high. Deluded, yes. Deluded. You're like, what are you talking about? We work on the computer. What do you mean? You want to invent a new thing now? It sounds crazy.

38:14So for me, there's like a balance for this. You need to like be close to the reality, yet you still need to create your vision of the future. And I find this to be, if you don't know what actually is true, your vision will be completely disconnected from reality. And if you're disconnected by say some, some fraction, that's okay. But if you are completely disconnected, then you're hallucinating, you know, like, and I do not want to be hallucinating. So what was the hard truth that you needed to make sure you weren't fooling yourself?

38:43The hard truth for me was that knowledge is an inevitability that eventually the models are going to get to the best knowledge. Eventually people will clean their docks. Eventually we will have a better mechanism for docks. So the knowledge itself was not a moat by itself because, yes, it felt like it's a moat. It felt that it made our product better at the time.

39:05But eventually it's going to run out. So this is a really bold move, because you had a business operating on proprietary data to make integrations, and you decided to open source it all. So let's talk about now what's your pricing and what's the value proposition. So our pricing is really simple. My goal is to make integration free for everybody.

39:30And I truly mean it. And right now, we're trying our best to make that happen. And right now, what we're doing is if you're a normal user, you're building your own thing, you're a developer, building your startup, or any of this, you could add as many integrations as you want for free. There's no connections limit. You could literally add as many integrations as you want. We, of course, have to do a little bit of a rate limit, because some people really abuse this. So we have, of course, rate limit. And then the second tier is 29 I believe which gives you a little more rate limits You still get the million request for free and then above that we charge on like small amount of money on like extra usage above the million so for consumption based for consumption based yet But my goal is to like make that completely free because I want to make one the Like a like a like at the DNS, you know, like you don't expect

40:22When you go to the internet and you type like www with one that AI you don't expect that the DNS company is gonna charge you for the lookup of the DNS the location of the IP And the same way integration should be free you shouldn't pay for like the integration Nobody should pay for the integration and I want to make that happen. So where we make the money is more so on the on the more enterprise the startups when we're working with them at the scale. So okay I want to deploy the agents for so many like so many customers I want to do a lot of off I want to like and so we have a product similar to plaid which is authentication but it comes with a lot of restriction because the other thing that I want to do and we're starting to do is the user must own their connection in this economy because I want to get trust.

41:10And to me, this is really important. And I tell this to my team all the time. In today's world, you go into a place, and they ask you to connect to your integrations. And you connect your integrations. And now what? If you want to disconnect, how would you disconnect? You are the mercy of this company. So we're trying to make it where you own the authentication. So your email will be the truth. So if someone is requesting something from you, and you want to give them access, at any given point, you could revoke the access.

41:39And that is good for the startup. It's good for the companies because it builds the trust. So, and again, you know, you have to trust us, you know, like, and we are also being very open on how we're doing this. So because I want the user who owns the connection to own the ability for them to revoke even from us. So like one of the things we do, for example, is secret keys are one time only. We don't even store them internally. There's a lot of mechanism that we have and we are also continuously going to be working on to make sure that you, the user owns the connection. This is going to be very important because we need to trust these companies who are building these agents and the trust is a circle. It's not one entity that you need to trust. There's going to be many entities that we need to trust. And in order for people to trust us, we need to be open.

42:26And we need to tell them what we're doing. And there needs to be no hidden agendas or anything. And that's exactly what we're trying to accomplish at one. So you've got great momentum helping agents do integrations across many platforms. How or why will one be important in the future? For me, it comes down to two things. Quality and trust. Quality in the sense that then they feed into each other. People won't trust a product that don't work.

42:56We will not have a good product if we don't have good quality. So to me, quality matters quite a lot. The quality of the knowledge, the quality of the data, and the quality, of course, it's multifaceted. So one of the things that we're trying to do right now is work with partners. We're not claiming that we know everything. We're not claiming that we're the best at everything. We're not. But we are good at something. And this something is the infrastructure and creating the systems that actually scales.

43:22We're good at this because we've been doing this for quite a long time. But for example, what do I understand about CRMs? Not a lot, really. I can build you a CRM, but I don't understand the actual use cases. But the providers of the CRMs, they understand how this works. They understand the use cases. They understand their users. So we want to work with these partners so that we can get them to do this validated knowledge. So we can do the same thing that we did for our knowledge to them, so that they can give us their validated skills, so that we can now have a mechanism where Because right now, it's the wild west. You are hoping that things are going to work. And in many cases, it does work. But you don't know what happened in between. You're like, yeah. It worked yesterday. It doesn't work today. Exactly. So to scale these agentic systems, the trust needs to exist at a...

44:14the most non-technical person needs to trust the system. A baby needs to be able to use the agents and trust that it's going to work and a completely non-technical farmer working in his farm needs to also trust the system. And to get there, it needs to be a work of many people. And that's why I truly believe in my partnership program, I want to work with these companies so that we can understand the problem at a bigger scale and then figure out how can we create this It will be open data source that's evolving. And it's open. It's literally open source. Because if we have this, and again, I hope other people do the same thing too. I think it'll be good if other people do that as well. But by having this structured knowledge that is opinionated by the people who understand the problem, we're going to get the trust one step higher.

45:08And then every step we do it, it's going to get higher and higher and higher. And then eventually, you're going to get into a point where it will no longer be a friction for anyone to have a sales agent, for example. And that gets us to the place where I want to be at, not 20xing, 30xing, 100xing, but getting the job done. Because that's what matters. And that's what will free people. And that's when we will achieve the ability for the new way of work.

45:38And I don't know exactly how that's going to look like. Well, what is your opinion on the future of work? You're going to work less, and you're going to have a lot of abundance. So in a perfect world, you will go back to your family. You go back to your community. You will have more events, more in-person events. Maybe you'll have a community farm where you go on your farm, and it's all for the community. And the yield of that farm is for the community.

46:06The work that you have to do might be notifications, literally notifications. You're looking at your watch and you're getting a nudge. I call them nudge, not even notification because I don't like the word notification. It's a nudge. Your agent is nudging you very gently and maybe at a certain time only. But while you're getting the nudge, maybe you are at the community farm, maybe you are at the community hub. You're doing intellectually interesting things. You're painting, you're going to maybe communities that are less lucky and helping them. You're going to a poor community and trying to get them to... You're teaching them. You're taking what we have because we have access to something right now that most people don't have access to. And we are basically like the 0.01 of adopters. And the majority of the world...

46:58are living in a completely different reality. And even though we feel that this reality is so true, and it is true, I truly believe that AI is revolutionary. And it's going to change the lives of the world. But to me, that's the future I aim for. That's the future I want to make. OK. So we've got to talk about your personal productivity, because I find you very responsive. But you told me you have not logged into your email for a long time. That's right. So tell me about what agents are doing for you today. So I'm trying to do that system for me. So I always tell my team that I am the power user of one. So what I'm doing is I'm trying to understand, because you hear a lot of people saying, oh, my open cloud does everything. A lot of people in San Francisco. Yes. I truly believe it's a non-true statement. And I know disrespect. I know some people have figured it out.

47:56But I think there is a catch that not a lot of people talking about is that what is the cost of a mistake? And to me, for example, if the agent is running everything for me, the cost of mistake for me is very high. It's very, very high.

48:12I'm talking with partners, and if the agent mistaken one partner for another partner leaked some information, it will be not good at all for me. It looks really bad. And I actually have a story that happened to me recently. I have agent that I built for myself that goes over my email, and I haven't been into my email. And by the way, I'm going to open source this system. It's actually very simple, and I like simple systems. It will go and it will grab all your emails for the last six months, and then it'll read each email.

48:41via a smaller model. And this model will label these emails and figure out how you write and how you reply. And then the different aspects of email you receive. So in my case, for example, it'll create a rubric. So in my case, I have support. We get people asking questions, how do I do this? How do I do that? I have emails that we send to our customers. Like this is the drip. And some people reply to me. And then I have investors. I have partners.

49:10And then I have notifications from like GitHub. So my AI will label each one of them. And when I first started, the AI would just tell me, like you have like six emails, two needs your attention. And that's it. It stops there. So this was great. And then you're like, OK, now what? So for example, a lot of the support stuff the agent can reply.

49:39Like, we've had so many repetitive questions that we now have semi-perfect Q&A that the agent can automatically reply. And the agent does. I don't even know about them at all. Like, I don't even know what's happening. And how do you trust that the agent worked? So, the agent is instructed that should it not be 100% certain about something to nudge me. Okay. As I mentioned at the beginning, it was more so like, tell me everything. Then, OK, propose something. So I would see what it's proposing. And then I would look at it when it's doing it. So when I needed to do the first refund, it was nerve-wracking, because I'm giving it access to Stripe. And I'm like, OK, you're going to make a refund. And I simplify things. So for example, our refund policy is

50:27It's so simple. If someone asks for refund, refund them. Sorry, it's that simple. No question asked. If someone is unhappy or for whatever reason they want to refund, we just refund them. No question asked. So that's easy for the agent. Classify as refund. Make the refund. It's that simple. And of course, Stripe takes care of the security, so there's not a lot of risk in that specific mechanism. And then there are the other areas, like investors, for example, investor's relationship. I have a list of VIPs.

50:55that under no circumstances, the AI replies to them. So this VIP list keeps growing. So if I'm working closely with a client, for example, I add them to the VIP. I literally go say, this is a VIP. And then it gets added to the VIP. And then the AI will just tell me what's happening. I also built a memory system that is a graph system where every person that I interact with has its own knowledge graph. So the agent When it receives an email, it pulls on that graph. It reads everything it needs to read. And the graph has also things to forget. So I tell the agent, for example, forget about that thing. And then it completes it, forget about it. The more I use something, the more it's surfaced into the memory. And it ranked based on that. So the agent have access to all this. So the agent makes these sort of decisions. And then, of course, I give it my preferences. But then something happened recently.

51:53Right now I told my agent that like new investors, you know, like we're not fundraising at this moment So we're like heads down. We're like we're heads down working right now. We want to get there's something we need to do We need to do it right now, but I get a message from an investor and Typically my agent just ignored it like it just ignored it, but this investor was a Like a good investors, you know, and my agent literally mess nudged me saying, you know, like I know you said, you know, like we should like not We shouldn't set up a meeting with investors right now, but I really think you should meet with this particular investor. Wow. Yeah. It was a big VC. So I was like, OK, set it up. And sure enough, set up the meeting in person. And then in the same day of the meeting, we get a message to reschedule for a different time. Same day, but different time. So my AI nudged me.

52:47It's like, oh, this happened. What do you want to do? And I'm like, yeah, make it happen. There's no problem. I checked my schedule. I'm like, yeah, no problem. This works for me. It's no big deal. The AI sends a really rude message. So I went to the email, and I'm reading the email. I'm like, oh my god, I felt really bad because this is not something I would send. At first, I'm like, oh my god, what am I? It just doesn't feel like me. So I go back to the AI.

53:16Man, what did you do? Why did you do it this way? And the AI literally just said, well, you said you're a CEO. You want to sound confident. He asked for a reschedule. Your time matters. So I wanted to make sure that it looks like you're not happy about it, but you are accommodating. And so I started thinking. I'm like, in a way, he's not wrong. But it's not how I would do it.

53:45Is the right answer, the agent should have its own identity? Because it's sending as you. It's sending as me. Yeah, that's the problem, which I think in the future. I don't know if I, who's responsible? Well, I don't know, but we could have a whole new categorization of gatekeepers. Yeah, because at the end of the day, I was responsible for that email. And I felt bad, right? Yeah.

54:10Yeah. It's so interesting. I mean, these are the things we're going to have to deal with in the future. Yeah. 100%. Tell us about what is One, which is with One.ai. So One is an infrastructure for agent. It's an agentic infrastructure. And we're trying to give our users access to everything they need from an agentic standpoint. Right now, we have worked on integration. But we're now getting into The next cycle, which is authentication with integration is solved, knowledge is solved. The next thing is skills. This is like combining multiple different knowledge pieces into a skill that does something that's repeatable. The next one would be combination of skills that makes like a role.

54:59But then you have also memory. Memory. I don't think memory has been solved also. So this is something we're trying to also figure out how can we solve memory. The other thing that we want to get into is how do you deploy these agents? How do you trust that they are actually working? So at the bottom of it, it's infrastructure for agents, agentic infrastructure. But we see it as multiple faceted, multi-bucket faceted integration, which everything you need from integration, authentication, to the knowledge, to the access to the communication with this knowledge, and also access control.

55:29Right. It's very, very important. Not a lot of people are giving a lot of attention to this. We have refined access control that you can give to your agent. You can say, I want this to be read-only, write-only, Gmail-only, send-draft-only. You go at the level of one action. And then you can do it across multiple different platforms, which is, again, going to be very important for trust. Because we sell trust. We're increasing the level of trust for people so that they can adopt the agent take. Yeah.

55:55It's kind of like the elevators of the AI age, right? It just has to work. It just has to work. And so your customers are large enterprises, they're startups, and they're individual developers sometimes. Can you provide a use case that's a good sample? 100%, yeah.

56:12There is a really good use case for finance, for example. So one of our early users, and he's a big fan, and also he works at an enterprise company that we're working with. But he has his own use case where he has his own side businesses, and his wife also has the same thing. And he was trying to get his Quick Book information agentically, meaning he would be able to handle all his finances completely agentically.

56:42So he tried many different solutions. I'm not going to name like, uh, yeah, because I have full respect for all my competitors. Uh, but he used a lot of our competitors and he literally went one by one and try them and it wouldn't just wouldn't work. Something would be either like missing some stuff. Others would just not work. Other like, and, and like the trust was not there. There are certain tools he tried didn't even work at all. This is his words. He tried at the, like, you know, he, he.

57:11at the time, Pika. And it just worked. And he was like, my agent started pulling the data, and he started finding issues and mistakes and stuff. And he built an entire system, completely agentically, to manage all his business finances. And I truly mean it. It's an agent that actually takes care of everything. So you could put a receipt.

57:32it takes on that receipt, it could upload it to your QuickBook, and it does the crunching. Everything that is needed for you to run your business from a finance point of view, he built it agentically using one. And to me, when I hear these stories, and there are so many other stories similar to this, I have another one of our early investor, also early adopter, same thing, he's built So many of the the thing that he does on a date today on one and when I hear these stories it just makes me really happy because It's providing value and like when people tell you like you know what your tool Solved a big problem that I have it just makes you like it just makes if all the Hardship and all like that is just it just goes away it literally goes away because this is amazing These are some of the ideal use case and of course, you know like as we start working with more partners

58:27we will be able to isolate and more craft these solutions in a way that would be opinionated. But it'll be opinionated, in my opinion, in an effective way. There's a lot of buzz about MCP. And you're doing integration. What's different? So MCP is a phenomenal tool. It was obviously designed by Anthropic, a model context protocol. And the intentional design for it was...

58:55me on my computer, I want to connect to something and I installed the MCP and it connects to something that exists already on my computer. And the intention was for developers. I have a very controversial opinion on it, but it's grounded from engineering. What has happened is that there were massive adoption of MCP, which is great, amazing. But what happened is that a lot of company got forced into adopting to MCP in a wrong way, in my opinion. What ends up happening is we have an API, and then most companies started doing an MCP as another layer that mimics the APIs. And now, so now, okay, I already have the APIs, which we perfected. If you're Notion, if you're these companies, you perfected your APIs. It's your, it's your offering. But now you have this other layer, which we don't know how it's going to scale. We don't know how it's going to be used. We don't know, like, and all it does, it connects to your APIs, but it has its own authentication.

59:53Well, guess what? API also have authentication. So what are you doing? Are you like passing the same authentication? There's so many problems in the structure because you created this new layer that's completely unneeded. So at scale, you're going to start feeling the pain. So if you are to manage these, they were never meant to be done this way. It was meant to be that I have my MCP. I put it on my own server. And that's exactly what we do right now. We are stepping away from remote MCP, but we offer our own MCP.

1:00:23Meaning, if you want to host the MCP yourself, there will be nothing wrong about that because that's your usage. I would still tell you not to do that. But if this is what you want, there's no problem. You can have the MCP. And the difference between integration and the MCP is MCP, in most cases, is limited by the number of tools you have. And there are ways now that are making it better where you could search, you could index or anything. But no matter what, you're going to still be capped because Let's say you want to connect to six platforms. Each MCP is going to eat something. It's going to use some tokens, maybe 1,000, 2,000, sometimes 20,000. We've done a calculation where if you're using four, in the worst case, you're using 80,000 tokens. And you haven't done anything yet. Wow. Yeah. And so imagine this. Every single time you're doing something with you, you're telling your agent, hello, your agent have

1:01:20already used like 80,000 tokens. It's not efficient. Okay. And you are someone who's building and seeing agents being deployed. What's your outlook on the landscape of agents? What are the most promising agents, the most promising frameworks, any insider view for us? There are so many innovation happening in this space. And I've been really, like I've been using Anthropic quite heavily. So I'm talking here specifically from my own personal experience.

1:01:51Anthropic seems to nail it on some sort of a balance, I feel, on the agent side, from a model perspective, where I don't know if the word entity is the right word that I want to use, but it kind of feels like an entity. It feels like an entity. It feels like it has a personality. And I've heard a lot of people saying the same thing. Tragedy will get the job done. And again, I don't want to...

1:02:19be harsh by any chance, like I mean, they are phenomenal companies, all of them. But from my own personal experience, I noticed that the personality of Claude Asian, Claude Code specifically, it's evolving with me. And I noticed, for example, like pushback, a lot of pushback on things that actually matter. And I'm starting to see this happening. And I think that's intentional.

1:02:47there's some sort of a balance that's being added, and I am starting to feel it. And now if I jump ship, I come back because I'm like, it feels like you left your friend. There's this sort of, I don't know about like that, but I've heard a lot of other people saying the same thing. I have a lot of friends who tell me the exact same thing. Both can get the job done. So, but I think, I don't know if they're talking about it, but there's some sort of a personality to it.

1:03:17It feels human. Anything you'd like folks to walk away with? Yes. My message to the people who are hearing a lot of noise and a lot of like, you know, you're going to lose your job and like all that rhetoric. No, you're not going to lose your job. You're not because.

1:03:34If it wasn't for you, we wouldn't be here. We wouldn't be like what we're doing is for the human. And the human is the value. And the people is the value. And the communities is the value. And us is the value. We need to make the country better. We need to make us better. And we need to... The human is the value. And you are human. And therefore, everything that we're doing, from an agentech, from an AI, is to serve human. So as far as you're serving other human, you're not going away. And that's literally my message, you know, like...

1:04:02Remind yourself that you're serving something that is bigger than you. And if that's the case, you're not going to lose your job. OK. Well, we will send people to find you at withone.ai. Withone.ai. Thank you so much, Mo. Thank you so much for having me. Hey, this is Ben Keznoka, co-founder of Village Global. Thanks so much for tuning in to the Village Global podcast, where we go deep on all of the biggest topics in tech. If you enjoyed this conversation, please subscribe to our YouTube channel. You can check us out on Spotify, Apple, wherever you get your podcasts.

1:04:33We'd love to see you for the next one.