Mistakes I made implementing AI | Innas Arabi

A Podcast | Innas Arabi

Pete Neubig: Welcome, everybody, to the NARPM radio podcast. I'm your host, Pete Neubig, and I have a very special guest today, Inaas Arabi with me. Inaas and I have known each other for many years. He is the COO at Block Realty, where he's spearheading rapid growth with operational excellence. And he and his team have a vision to double the company size within the next 12 to 18 months. He brings over 25 years of leadership experience across the real estate and prop tech landscape, having managed and scaled operations at both institutional and mid-market levels. His background includes pivotal roles at American Homes for Rent, AltiSource, Zillow, and RealPage, where he previously led product and industry strategy for Propertyware, which is where he and I first met. Inaas, thanks so much for being here.

Inaas Arabi: Thanks, Pete. It's always a pleasure to join you and have a conversation with you. So I'm really excited about this conversation today.

Pete Neubig: I'm excited about this as well. Inaas and I were having a conversation and I was telling him about a webinar that we would put on about the top 10 AI tools. And he goes, oh, man, I want to talk about AI. And I'm like, hey, let's just do a podcast on it. And so here we are today. So I wanted to talk. I want to ask you a couple of questions, Inaas, right? So first thing is, why is AI so important for property management?

Inaas Arabi: OK, so, Pete, I appreciate you so much for putting this podcast together and talking about this important topic. Look, I want to make a couple of very quick points to make sure that people are understanding where I'm going with this, because this is really important for a framing perspective. Number one, I recognize that our business is a relationship business, right? So when we talk about AI, we're not talking about replacement of relationship with technology or with AI. But what we're really talking about is the ability to be able to build for scale and for growth and obviously for operational excellence, which allows you to have much better relationship or build much better relationship. Right. So I want to make that very, very clear up front. And then number two, what I also want to make very clear is that I do not have any specific support or a sponsorship of any sort from any products or anybody that we're going to talk about today. This is purely from a personal experience as well as personal need and personal wants. Right. So having said that, let's talk a little bit about where we're at in the industry. And I'm actually going to say this is at the level of what we call in the IT industry Code Red. You may have heard the term before. You may have seen some of the news. A lot of times you get ChatGPT going out saying, hey, we're issuing Code Red, then followed by Google says, hey, we're issuing Code Red and all that. What does that really mean and what does it mean for our own industry? Code Red in IT typically means I'm dropping all of what I'm working on to concentrate on this because I see the value of it. This is the thing that it's either going to make or break my business, my industry and what I'm working on. And that's why when ChatGPT comes up with a new model, usually Google gets notice and they say, hey, I'm issuing Code Red so I can put a lot of resources against it and vice versa. And they play this game. Right. So I am saying we're issuing or I am issuing a Code Red for our industry. It's really, really important for us operators to learn and understand the technology and how do we implement it into our industry and our day to day operation. The problem is that all along we've allowed vendors to control the messaging. So what do I mean by that? The vendors that they, so to speak, they sell products, some AI, some not, they're the ones that they're they're communicating to us about the importance of AI and whether it does or it doesn't do what we want it. The truth to be told, that's not the right messaging. The messaging should come from within us, the operator, people like yourself, people like myself, people like all the ones that they're running companies, because we're the ones that we know what can we do with it and what we can't do with it. And also what is smoke and mirrors and what's real. Right. I'm also going to tell you, and obviously goes without the description, that the market is getting a little bit tougher for all of us on all fronts.

There is a little bit of a downward pressure on fees and on actual management costs and fees and all of that. You may have seen some of the new regulations coming on in some of the states where, like, for example, if you're using a benefit package, you would have to put that very clearly and describe it very clearly in your listing descriptions. And some places you even have to add it into the price. So people are aware of that. All of these regulations is going to put a lot more downward pressure on your income, on your revenue. At the same time, there is also an upward pressure on your cost because your biggest cost in our industry is what is human beings, is people. Right. And I love them to death. But we know whether it's local or whether it's remote, it's getting a little bit more expensive by the day to be able to find the right talent. And that is pushing the cost a little bit higher and higher every day. So the business owner seems to be squeezed between a downward pressure on on revenue and that upward pressure on expenses, which leaves you with even less to be able to make this work. And let's face it, our business is a very low profit margin business anyway, because it's all based on relationship. Right. And it's got a lot of payroll and HR costs to it. So the only saving grace for where we're at now is to use the right technology to be able to get better results. Now, you can't just open up the book and start just throwing technology at it right and left. That's also not going to work. You really have to be very methodical and purposeful about where and what you want to do with it. I know it's a long answer to your question, but hopefully I framed it enough to make it, again, a code thread for our industry to really, truly make sure that we are paying attention to it. And we're putting a lot more thought process around it and working toward building something with it.

Pete Neubig: Do you think because I I'm in my 50s, I'm kind of resisting the whole AI thing. I use it, but I'm not embracing it. Right. Because I have to use it. But all the stuff that I'm reading about it, it's not really where we thought it would be. Like when I first came out, I thought it would be more advanced than it is today. Do you feel the same way or do you think it's advanced, you know, more so than what we thought?

Inaas Arabi: OK, so you would have to look at what you're saying and put it in the right context, right? What are we talking about when we're saying advanced that we're saying where AI is going to rule the world? Certainly not. And it's it's not going to be there anytime soon. But is the technology itself getting better by the day? And the answer is absolutely. Some of the things that that there are available to us today were not available last year, and those things are are purely amazing. So I'm going to give you a couple of examples, right? Typically, when you go to a website, you open up a browser, you plug the URL in, you go to that site, if you need credentials, you plug in your username and password, you log in, you go do what you need to do, you finish out, you log off and you're done. Right now, imagine. Imagine an AI bot or a browser extension that is doing that on your behalf. Using your own credential, the exact same steps that you're doing, the exact same push buttons that you're doing yourself to be able to produce the same results that didn't exist a year ago. It did not exist at all, right? The technology has gotten so better at it today. Additionally, what I think it's also important is the technology itself started to understand the limitations of what was wrong with it initially and started to fix it. So let me explain what I mean by that. You probably have gone on to ChatGPT and you plugged in questions, right? And you got nice answers and you worked through those and everything is fine. And then you came back to it a week later and you asked a question about something that you've done a week prior. And sometimes you may have gotten an answer that was half assed. Sometimes it was really good. Sometimes was just it looked like that it's not remembering what you were saying and what you've talked with them about with the model about. Now, this is we're not talking about the systematic settings that you set yourself up for because those settings are going to be available all the time.

We're talking about the actual conversations back and forth, right? So if I had gone on to ChatGPT and talked about a trip that I'm taking to Europe and I came back a month later and I started asking him questions about the same trip, most likely a few months back, you would not have gotten any answers on that because there is a degrading memory for what's going on on the technology today. There is a the right technology has enough memory behind it that actually remembers the context of what's going on now, why that's important for a business like ours. It allows you to be able to build upon all of those things that you've had to actually make sure you maintain the relationship. So let me put this in perspective for you. Let's say if you have a AI chatbot that or agent that is doing your email for you, and if that email is not realizing and recognizing all of the communications that were done with a particular owner a month or two or three down the road, they may write a email that is contradictory to anything else that you may have wrote before. And the reason being is because there is no memory. There is no contest for that technology to kind of keep up with all of what's going on. Now, that technology have gotten to a point where are some products out there close to infinite, close to infinite. So it's not infinite, but close to infinite memory. So what that means, it's going to remember every single thing that you're doing with it. Every single thing that you're doing with it. Now, imagine that for a second. You and I, we built relationship over multiple months, multiple years, right? But when we get together, when we talk about something, we remember, oh, yeah, I remember when we talked about X, Y and Z a few years back, you remember what we did? Remember this, that? Imagine the same experience with this technology. So I would tell you today, the technology have advanced itself a lot more than we all think. But you have to recognize what that really means and what's the applicability of it for our industry. And those are some examples that I'm illustrating for some of our industry to be able to understand where we're at and what's going on with it.

Pete Neubig: And where is Block and Associates using AI today?

Inaas Arabi: Yes, so we were one of those companies. I have a funny story for you that it's worth saying. About maybe 18, maybe 16, 17 months ago, AI was not necessarily on our radar, to be honest. It was more of like, hey, it's a novelty. Everyone was like talking about cheap chat, GPT. We were going in there and kind of playing nice and asking questions and we getting answers and everybody was all right. Right. I had a coffee with a friend of mine who leads the marketing for a very, very prominent brand. Really prominent, actually, like we touch it almost every day, all of us. Right. And he was talking about the way that they're using AI in their business. And he mentioned a couple of things that they were like quite interesting to me, and I was like, wow, this is I've never heard of that, never seen that. This is kind of interesting. So I started doing research. And once you start doing research and you start understanding what you're working with, you recognize the importance of being on the forefront of this, not on the back front of it. So just like every business, we called the vendors and we said, hey, what do you have in AI? Bring it on. Let's take a look at it. Let's play with it. Let's try it out. And we've tried several things in 2025. As a matter of fact, I'm going to tell you that we've had three or four implementation of AI products in our business. Some failed miserably, and I'm happy that they failed because they taught us quite a bit. Some succeeded somewhat, failed somewhat, and some succeeded to teach us what we need to do and what we need and how we would need to set things up. So having said that, let me tell you about the installation, the worst one that we've had.

Pete Neubig: Before you do that, what made you say you didn't read, you were doing research, you called the vendors, was there like there's different areas of the business that you can implement AI. What made you choose, like what area did you choose first and why did you choose that area?

Inaas Arabi: Yeah, so where you would want to get AI is in places where you have repeatable actions that may not necessarily be a relationship builder actions. Right.

Pete Neubig: So like maybe leasing at this point?

Inaas Arabi: That's one of them. Right. I'm going to give you one that is really probably the most simple, which is the front end front desk phone number or a replacement of IVR. Right. A lot of companies.

Pete Neubig: Yeah, once you get that phone, they just hang up. Yeah. You don't want to be in that phone tree.

Inaas Arabi: Well, either that or they get frustrated and they just start pushing buttons.

Pete Neubig: Or hit the zero button.

Inaas Arabi: Yeah. Yeah. One, zero, anybody that answers anything. And then and then you hear the typical, well, I have not. You know, you guys are too hard to get a hold of. Right. And a lot of it, a lot of it could be very simple questions that you get all the time. Right. Like one question that we get all the time. What's your address? Right. Or what's the email address that I send your invoice to? Or, hey, how do I how do I apply?

Pete Neubig: Yeah. Basically, 80 percent of the questions are stuff that is repeatable.

Inaas Arabi: 100 percent. So what we thought initially is exactly what you're saying, Pete, we're like, hey, leasing is a really good area. Front desk is a really good area. And then we had another thought and I'm going to talk about it here in a moment. But let me start with the leasing, because that's actually the one that we started with.

Pete Neubig: You did. OK.

Inaas Arabi: And if it yeah. And it failed miserably and it failed not because the technology really failed because of us. But let me kind of outline that a little bit more. So we partnered with a product or a vendor that came in and says, hey, I'll build you a chat bot that sits on your website, as well as a voice agent that is going to take the calls and is going to try to explain to people about what you have and rental listings and try to set them up for a showing by pushing them onto the showing site so they can schedule themselves for a showing or if they want applications, we send them the link for applications, things like that. Right. The general examples of what happens. OK. We did deploy it. Failed miserably for three weeks. I've had more upset leasing agents that I could ever count in my life. And that was a really good moment for us, because what we've learned, the first thing is the change management. It's not just with owners or with clients or with vendors or with customers. It's actually also with your own team internally, because if they don't understand the technology, what is it doing? How is it going to help them? How is it going to take them to the next level? They're going to be very negative about it and they're going to be not willing to participate or support it. And if they don't participate, they don't support it, it backs fired on you and then you lose that.

Pete Neubig: Yep. Additionally, you got to show them the vision on how it actually helps them, because, look, a lot of people are resistant to A.I. because they just see it as a job like, you know, virtual assistants. Five, seven, eight years ago, my team was resistant to virtual assistants because they thought they were losing their job. Well, now the A.I. is they're resistant to because they didn't lose their job. And the other thing that I feel is like, you know, a lot of people think I buy A.I. I, I get the company to kind of build it for me and then I just turn it on. I think A.I. is actually like another team member. You have to train it, nurture it. There's got to be, you know, there's got to be like a way to do escalations. And so if you just think I'm going to turn this on, just like if you hire somebody and you don't train them, they can't really help you. Well, the A.I. is kind of the same thing, but it could be exponentially worse because it's sending out stuff to your clients, your vendors, your in this case, your leasing agents. So what lessons did you learn on this on this faux pas?

Inaas Arabi: We it's so beautiful the way you put it, because that was one of the lessons that we've learned, right? Like you've got to be able to train it and you have to give it enough time to be able to train itself as well. And what I mean, train itself, the phrasing of this is an additional team member is probably the most accurate phrasing ever, because you have to treat it almost like a new member to your organization, to your organization. So you got to provide it the right documentation, access to the information, give it enough time to train. And you have to train it yourself by telling them all of the specifics. So like one of the things that we did not do very well is provide the documentation in a very simple format that the bot could understand. Right. And we didn't really provide a good access to what we call like the listing widget. So what happens? The bot sat on top of the website. Right. And remember what I said to you a little bit earlier, you get a little bit of a degrading memory every once in a while. So what happens if it's reading your site, right? The page, by the time it gets to that last listing, it's probably started to forget about the first listing that you've had because it doesn't have a direct access to it. It's just reading whatever you have on the website. Right. And if every time they get a call, that bot has to basically do the same thing all over again, then they start missing specific details about listings because they're not getting all of that information put into their memory for a long term. The only way to fix that is to provide it access to your listing information and put that into the memory of the bot. So this way, either the voice agent or the bot is able to actually access it every time instead of having to read a website to be able to tell you what's going on.

Pete Neubig: How you how you gave them access to the data was a learning.

Inaas Arabi: Absolutely.

Pete Neubig: I wouldn't even thought about that. Yeah.

Inaas Arabi: Yeah. How to give them access to data. We were again, we did not do change management for our employees. That was that was a miss. Definitely. And then three, we didn't give it enough time for it to kind of fix all the issues. Like literally it was a knee jerk reaction. And I, I, I would say that the I've received a lot of negative feedback that I was compelled to having to take it off because I just could not deal with all the negative feedback from the leasing agents.

Pete Neubig: Yeah, it's garbage. Yeah. Right. If you if your data is not good, it's just going to exponentially get that data out to more and more people. And it's not the right data. So precisely. So lesson number one is have access to data a different way. And lesson number two would have been to verify all your data prior to. Would that be something you have to like? That's a big that's a big project, right? Because you have to actually go through all because, you know, especially if you're in business a long time, you got a lot of stuff out there and stuff. Sometimes we're supposed to sunset it and we don't. Right. Like, oh, we don't do it that way anymore. But if it's still there and the A.I. has access to it, it's going to think we do it that way, correct?

Inaas Arabi: Yes, 100 percent. 100 percent. So the knowledge base is is like a a must to make sure that you have it right. Now, the knowledge base on the leasing is kind of interesting, right? Because what the knowledge base is technically the information about your properties, that they're listed for rent information about your application process, information maybe about like what your moving process looks like. Maybe information about what? You know, like how do they go about showing or doing something within the process of the leasing? So it's not necessarily going to have to have all of your SOPs for the entire company. But if you if you built the right knowledge base for the right skill set, for the right job, for the right technology, then you would win because you gave it the information.

And there is one more last thing that was absolutely massively important to keep in mind. Off ramps. So let me explain what an off ramp is. Remember what we just talked about when you call an IVR and you get frustrated, you stop pushing any buttons to be able to get anyone that will get to answer your call or are you going to hang up? Right. We all do it.

Pete Neubig: Are you on the phone? Operator. Operator. You look like a madman. Somebody says to me, my card, I'm an operator.

Inaas Arabi: Hundred percent. Right. And then but the problem, if you do not have an ability for somebody to be able to connect with a human being, it even make this whole experience much, much worse. And people get totally frustrated. That makes sense. So one of the mistakes that we've done earlier on is there are no off ramps. So like if somebody is wanting to talk to someone. Right. Yeah, you got to give them an you got to give them an opportunity to be able to do it. OK, that's a that's a very good point. And I'm going to also this is a nugget for everybody who's here in us because this is from experience. The relationship between the people and the technology is predicated on multiple different things, is multiplicated, predicated on the location of where those people are, predicated on their knowledge base of the technology. And sometimes their age group or whether they're trying to do. So when you think about that, because you don't know all of those details for every caller that you're getting or every person that you're getting, you have to build an off ramp for us where we're at. We're in North Carolina. The average age for our folks are probably a little bit higher than some other areas of the of the country. Nothing wrong or right about it. It's just the facts. Right. We're not as technology advanced as like the Silicon Valley or some parts of California or New York, where we're risk averse as far as the type of people that we deal with. And when you put all of that together, our research showed us that about 50 percent of our people are willing to go through the AI experience and 50 percent they wanted to talk to a human being. OK, if you know that and you understand that, then you got to give them an opportunity to talk to a human being, because you're what you're going for is the saving of that 50 percent, not 100 percent. Right. Meaning like when you're building these products, you're not trying to build a product that is going to solve 100 percent of the cases because it won't if it's if it's customer facing. It's only going to do about 50 percent of the cases, at least in our example, in our experience. So initially, when we did the leasing, we were we were not very progressive enough to recognize that. So we did not have an off ramp. And that caused like quite a bit of havoc for everyone involved. You know, it really it caused like a lot of a backlash on us. So anyway, so those are the lessons. Number one is the change management. Number two is recognize the knowledge base. Make sure that you treat any of this technology as an additional member to your team and work through it that way. And then finally, you have to be cognizant of what the technology is able to do and given people an off ramp to talk to a human being when they wanted to or get to a point where it just cannot respond.

Pete Neubig: I would even say like there's some prerequisites here that I'm hearing you say, but not really saying it like pointed. Right. So one prerequisite is you have to understand your process really well. Oh, yeah. You don't understand the process and you implement AI chaos ensues, right? So you have to like if you have to hire a black sheep global EA consultant, whatever it is, or do yourself, you have to understand your process, right? Then, too, you have to review all the data that this AI is going to have access to make sure it's up to date. Would that be a correct statement? 100 percent. 100 percent. Absolutely. Then three, you have to get your team buy in. Right. You got to show them like, hey, instead of you working six days a week to 14 hours a day, we're going to get you down to five days a week, six, six hour days, and everybody's going to keep their job and we're going to be able to grow and we're going to give you all these extra benefits. Right. Something like that. Right.

Inaas Arabi: 100 percent.

Pete Neubig: OK.

Inaas Arabi: Yes.

Pete Neubig: Then you have to design how the AI is going to access the data. So that was a big miss. I never I never heard that before. So if you listen to this, that's a huge that's a huge catch. Then working with the vendor to build out the AI and how it's going to how it's going, how that team member is going to run your process. Then you have to train the AI. And I'm guessing that training can take anywhere from 30 days to 60 days or something like that. The longer, the better.

Inaas Arabi: Depending on the complexity of where you're interjecting that AI. So like if you're putting it in maintenance that learning, it's probably more than like 90 to 120 days. And it never really stops. Right. Because you're always going to get new cases, new things that you got to be able to not redo the process, but maybe put more knowledge base into it to make sure that you've got all of the specifics of what needs to happen. And the only way to overcome that in the beginning is make sure that anything that it's off base from what the knowledge base looks like, that it comes back to a human being to be able to manage. And then that human being provides the learning, the continuous learning to the to the technology, to keep it up and running.

Pete Neubig: And are you running AI in conjunction with the way you normally do it before you actually give it more like, yeah, freedom or flexibility? So talk a little bit about that, because we're running out of time. But I think this is really interesting to like, you know, how like now we got it. We have the team buying. We got implementation. But how is it a turn on, turn off thing? Or is it can I turn it on?

Inaas Arabi: And like, no, a little bit about you never you never want to do a turn on and turn off. That's probably the worst mistake ever. I think it's concurrent. So in our world, we were really lucky to to have different offices. So what happens? We basically the second installation that we've had with the I we used one of our offices is like a pilot as a testing ground. Right for that particular bot. And then the other one didn't. And then we can compare and contrast results, feedback where we're at, where we're missing, where we're not missing. And that allowed us to really fine tune what's going on with what's happening. And we also could find out if there is a mistake somewhere somehow. Right. But it's never a turn on and turn off. It's really should be more of a concurrent. So I would do like a pilot on a certain thing, even if we take leasing as an example. Right. So maybe you do a pilot where the AI gets some calls and your human being gets some calls and you compare and contrast the results. Right. Where are, like, what am I seeing? Am I seeing a lot of people, you know, pressing that zero and they want to talk to a human being because maybe there's something off in the language, um, are they getting a little bit turned off of some of the experience?

Pete Neubig: Um, like you may have seen AI is giving them the right data. Do you have to actually listen to the calls and just to make sure, or are you guys testing it yourselves and making the phone calls into yourself.

Inaas Arabi: That's the beauty about the technology. And that's the beauty about what happens. It basically transcribes all of the calls and it provides its own feedback if you want it to, or you can build another agent that does the feedback for you on all of the transcription, right? Because what happened, if you're, uh, and I know it, it, it sounds funny, but it is.

Pete Neubig: AI is big brothering the AI.

Inaas Arabi: Well, but you know what, it, when you think about it, it's all researching data and telling you whether it matches or it didn't match. Right. Um, but it, it's garbage in garbage out. So if you start off with the right data and the right processes, and then there is something off that is going on, that's where you need to fine tune. But I'm going to tell you the technology have gotten good enough that you can basically tell it today not to use anything else, but the information that you're providing, um, you can also do what we call a full safe. So like, for example, what happens when there are no knowledge, nothing, um, either goes to a human being or if it could access a, uh, an LLM model, um, large language model to be able to come up with an answer, it can do that. But I'm going to caution everybody here to say that there is hallucination that happens all the time. And if you, if you've never seen it, um, you know, try it with ChatGPT. They could write you an email with a full legal language about certain things that actually doesn't exist. So the hallucination does happen and you need to be cognizant of that. But if you, if you only built the technology that uses your own data or the data that you verified, like for example, um, some people might have a depository of all data, like they use notion or they use Google docs or, uh, Google G drive or something like that, or even the Microsoft, uh, you know, um, one drive, all of those are just depository for all your data and your data is typically written in some kind of formats. So if you provide that to the bot, then that's what the bot is going to be using or the technology that's, what's going to using, if you provide the right instructions, that that's what you'd want it to use, but the knowledge basis is like one of the premier thing that we look for when we build any technology today with AI.

Pete Neubig: Were, were you able to build, like if somebody was talking to an AI and they said like a buzzword or like it had to be an escalation, were you able to escalate that out? Uh, like, would that be an off ramp that, uh, that you just described earlier?

Inaas Arabi: Exactly. So this is two things. Um, you have to put rules and those rules are, are, uh, what we typically call like transfer rules. So like, for example, if somebody calls and says, I'm getting my attorney. Involved.

Pete Neubig: Yeah. Oh, let me get you over to Pete and then my phone's ringing.

Inaas Arabi: Right. It's basically going to go to the compliance department. If they're talking about accounting, it's going to go to our accounting department. So it's really, it's more about like figuring out where can I go with that particular thing? Now the off ramp happens regardless. Meaning like somebody in the middle of the experience says, okay, I don't want to continue with this anymore. I want to talk to a human being. Then we would need to pass them on to a human being. So like in our example, if you call our front desk and you happen to get the AI, because sometimes you might get a human being, sometimes you might get the AI bot. That AI, um, in the middle of the conversation, if you say, I want to talk to a human being is going to say, wonderful, fantastic, please allow me to transfer you to a live front desk individual, and then they're going to go through two or three people on our team until they find somebody who's able to answer and then you pass them on.

Pete Neubig: Yeah, exactly. That becomes a terrible experience. So you got them right in the right spot. This is fascinating and awesome. And I know you and I talk all the time and I could probably talk to you another hour, but we have to be respectful of everybody's time. So, uh, I just want to say thank you so much for being here. Appreciate it. And I think we need to get you on and talk about when, when you had the positive AI, right, we talked about all the negatives and we'll talk about the positive. Maybe we'll get you on again. And we just like, here's how, here's how an successful AI implementation look like.

Inaas Arabi: So, and go ahead, rest, rest assured. I am extremely bullish on it. I'm not, this is not supposed to be a negative thing about AI. I'm actually extremely, extremely bullish. And I'm going to tell you, you know, that we're very well known for our pod system. So the next evolution for our pod system is we actually going to get a AI agent included on each one of our pods. It's that important for us. And it's that, um, extreme of, of positivity as well as bullishness on it. So even though I've talked a lot about the, the, the lessons learned, but this is not to persuade people not to do it is actually more of the opposite. I'm trying to teach everyone and tell everybody what some of the mistakes that we've done earlier on, so they don't make them anymore.

Pete Neubig: You're like, you're like maybe a few, uh, you know, you're a few yards ahead of everybody else and you're pointing out the potholes and this and that, right, you're just pointing it out. Like you are probably ahead of most people. And so this is great for them to, to listen to, to this podcast and be, um, you know, kind of, we laid out really a great, um, kind of like roadmap for them on what to be, be wary of this way. They appreciate that better experience than you did. Right.

Inaas Arabi: Yes.

Pete Neubig: What's the, uh, smart man learns from some, some other, uh, smart man learns from their own mistakes, but a wise man learns from other others.

Inaas Arabi: That's a hundred percent correct. And I would love to be able to, to be a, um, a helper in, in some way, shape or form to be able to provide that feedback, because I do believe all of us are into this together and we have to be able to do this right. And, and I'm going to say, we all need to figure out a way to be able to implement this technology correctly into our businesses, but there is plenty of use cases that you can absolutely do this with. Um, and I'm happy to come back again and talk more about use cases. Cause you'd be surprised where we, where we plug this in and what results are we really getting today.

Pete Neubig: If it asks, if, uh, if somebody wanted to reach out to you, what's the best way to, to reach out to you?

Inaas Arabi: Um, LinkedIn is usually a good place. Um, you know, inaas@blockrealty is usually a good email address. Um, anyone that, that wants to talk about it or have a question or whatever, look, uh, I don't have any product to sell. I'm not a, um, asking for any consultancies, but I'm happy to help whichever people that they're trying to accomplish things with this AI technology. And, you know, I consider myself a little bit, like you said, few yards ahead based on the research and the work that we've done, and I would love to be able to help people not go through the same mistakes that we've gone through. And, you know, get this technology to be implemented and deployed quickly and provides the best impact for their businesses and themselves.

Pete Neubig: I'll say this, this industry has a lot of people who give back, but no, no more than you, man, you are one of the top people out there and you are, you're such a giver back to, uh, to this industry. And, uh, I appreciate that. If you are, if you are listening to this, I promise you, AI is not going to take away your virtual assistants. You're going to need them to run your AI bots. So give us a call at VPM solutions. Uh, actually just email me direct pete@vpmsolutions.com. And if you're listening to this and you're not an NARPM member, shame on you, NARPAM.org, or give them a call at 800-782-3452, 800-782-3452. Inaas, thanks so much for being here, buddy.

Inaas Arabi: Appreciate you, Pete. Thank you. Thank you.

Pete Neubig: See you, bud.