Could AI Actually Replace Your Entire Call Center?
What if you could cut call center costs by 90% while providing better customer experiences? In this episode, we explore how AI voice agents are revolutionizing customer service and sales.
In this episode of Leveraging AI, Isar Meitis sat with James Lindsay, the CEO of LogicLabsAI, helping businesses implement AI solutions for call centers. With over 15 years of experience, James provides insights into this rapidly evolving technology of conversational AI.
Here are just some of the topics we discussed
AI News of the week:
About Leveraging AI
If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!
Hello and welcome to Leveraging AI. This is Issar Meitis, your host. One of the toughest jobs you can have is doing outbound cold calling sales. And the reason it's tough because the person on the other side is not necessarily wanting to buy from you. You're approaching them and it's. As close as it gets to being a spam caller, only you're calling about something that they may actually want. The flip side, from a different perspective, one of the most frustrating things as a consumer is customer service and especially customer service with large corporations. You can never get the right answers. It's never the right person. They keep on moving you around from one person to the other. You have to retell your story again and again and again, despite the fact they quote unquote have notes in the system. And it's just a big mess. Wouldn't it be amazing if there was an AI solution that can solve both these problems, that can actually do outbound sales better than a human, and that can do customer service while being connected to all the data and resources and decision making process that can actually provide effective customer service over the phone. Well, this sounds like science fiction and something we may have one day, but it actually exists and working extremely well. Today. And this is going to be the topic today's podcast. And while you may not think this is applicable to you because you're not in that kind of a department, this can replace any communication with clients, including text and voice. So if you're in business, you have communications with clients. Otherwise, you're probably not in business. So this is probably applicable to you, and I think you'll find this fascinating At the end of the episode, I'm not going to share a lot of news. I'm going to share one piece of news, but it's so important that I'm going to invest a couple of minutes to explain to you what's actually happening and why is it so important. So stay tuned for that. This episode is brought to you by the generative AI business transformation course for business leaders that is brought to you by multiplai. ai. There's going to be a link in the show notes, so you can open your phone right now or your computer and just click on the link and go there it's a fantastic course that has helped business leaders transform their business And help people completely transform their careers. And now let's dive into what the future of call centers and contact centers looks like with AI. Hello and welcome to Leveraging AI, the podcast that shares practical, ethical ways to leverage AI to improve efficiency, grow your business and advance your career. This is Isar Meitis, your host, and I've got a fascinating show for you today. And at the end of the show, I think you will get to the same understanding I got to when I started exploring this topic, which means the days of call centers are numbered. And maybe I'll broaden that the days of contact centers are numbered because AI can now not only send emails and do chats, it can actually have live phone calls with people and it sounds completely human and it can run through sales calls inbound or outbound and customer service calls and so on. It can run 24 seven in any language and at a fraction of the cost. Of an actual call center. And the other thing it does it at a very high consistency, meaning all your agents will get to the same results. So just think about how much time and budget and efforts and resources go into contacting and following up with prospects and with clients, either for sales or for customer service. And once you put a eye into that. Your focus, the human focus can be on closing the bigger deals on resolving the extreme customer service scenarios. And everything else is almost on autopilot. I guess today, James Lindsay, he's the CEO of, Logic Labs AI, who is, which is a business that's helping other businesses implement these kinds of solutions, meaning replacing entire sales and customer service processes that include every aspect of that from understanding where the contact is coming from, where it needs to do, what's the flow, chats, emails, and yes, also live phone calls. So this sounds almost like science fiction, but it is science and all fiction. And hence, I'm really excited to have James as a guest today, James, welcome to Leveraging AI.James Lindsay:
Thank you. Isar, I appreciate it. It's a pleasure being here on this podcast and I've been enjoying many other episodes you have. So I can't wait to get into the details of today. And it's truly an exciting field that we're involved in.Isar Meitis:
So I want to, before we dive right in, I want to make it a clarification from you so people understand we're talking about actual phone calls, right? People, one human on the other side and an AI on the other side, right?James Lindsay:
Yeah, exactly. So the days of auto dialing and people picking up and there being a huge pause and saying, hello, this is so and so is this, are they there? And it's a recording. these are real life calls when someone picks up on the other side, you really can't even tell it's not an actual human. Because of the technology that's out there and, and the way that it speaks to you with inflections and answering questions and even trying to trip it up. So it's pretty, pretty incredible.Isar Meitis:
So before we continue and dive into the how and what and why this is all done, let's listen to a snippet from one call so people that have never experienced this before can. really understand that this is real and it's happening right now and it's happening successfully at scale. So it's not tested in a lab. This is out there in the wild, actually making phone calls to people. If you can give us a little bit of background on what the call we're going to listen to is about, and then we can listen to the call and then we can dive into, first of all, what's the business impact that this is having the businesses that are implemented this, and then how to actually, put something like this in place.James Lindsay:
Beautiful. Thanks. Great. Great concept to move forward with. So the call will be listening to is specifically in regards to an insurance agency. So insurance agencies always have policies that need to be renewed. It's very labor intensive. Sometimes they get skipped because they get lapsed. And so in this particular call, you'll hear, if this is an outbound call that was made And it's to contact someone who has a policy that's up for renewal, and they want to go through and make sure that it's the same coverage, the same, policy that they wanted, or did they want to switch it? And so they, the conversational AI is able to do all this. it's not robotic. It's actually able to make decisions, to steer a conversation, to take orders, or even pass it off onto, say, a senior account executive if needed. That's the call we're about to go through and listen to. So I guess we'll do that right now.Isar Meitis:
Awesome. Okay.James Lindsay:
This is Sarah from BarryIsar Meitis:
Jenkins Real Estate Group. How are you today? I'd be a whole lot better if I wasn't getting these scam calls. I totally get it. Those scam calls can be really annoying, but don't worry. I'm not here to scam you. I'm actually calling because it looks like you opted into one of our ads about looking for homes for sale recently. Can I talk to you about that? Yes, but you're calling from New York and I'm in Kentucky. Oh, sorry about that. Our team is based in New York, but we work with clients all over the country, including Kentucky. So no worries, we can definitely assist you with your home search. Can I ask what area in Kentucky you're looking for? Horse Cave, Kentucky. Got it. Horse Cave, Kentucky. That's a great area. And do you have a time frame for when you'll want to purchase a home in Horse Cave? I'd like to purchase it as soon as possible, but the down payment is what's killing me. I'm on a fixed income. I don't make that much myself. Of course, it's for me and my family. There's five altogether. I understand that the down payment can be a challenge, especially when you're on a fixed income. Don't worry, we can definitely help you explore, um, different options and resources to make the process more manageable. Have you spoken with a lender yet? No. James, this is insane, right? So we're implications of what we just heard that again, this is not a test recording. This is not an example. This is not done in a lab. This is an actual insurance company actually applying this. I want to go back to what I said before we said we listened to the call. What are the business outcomes that your clients and people that are using this technology are gaining from using thisJames Lindsay:
technology? Yes. there's many of them, but the first one, let's talk about the financial impact that it has. Cause that's everyone is concerned with. Okay. At the end of the day, we as a business, you need to make money. So how does this work? So just specifically when you take out the human nature of what happens, in a standard phone call, say as a regular human agent is able to do about 50 phone calls a day. If they were selling a product that say, had a value of, of$500 on that particular one or say, Let's say it's$250 on that particular one. They would close 10 percent of those. So you're at$2, 500 for that particular. By using the conversational AI, and this is just with one agent and understand that these can be stacked. If someone really wants to go crazy, they can stack these and have it to 100, 000 calls simultaneously, but realistically for most businesses, if you could 20 X that number and say the agent, the conversational agent can go and do 1,000 phone calls a day and have the same percentage. They're probably higher, but the same percentage of 10 percent on that you end up with at$250 at$25, 000 that they closed that day. Or if it's a$500 value of product, they close$50, 000 a day without a human, just using the conversational agent. So right there, the impact financially is tremendous. But then you're probably asking the question, okay, so what is this going to cost us to keep this motor running this operation running? How much is conversational AI? And that's even when it gets better because, you have the cost say for the insurance appointment that was closed. On average, that is$25.98 on average. That's what is averaged out across the board. if you were to have an actual human or a person trying to close that sale. It would take multiple days, maybe multiple times. And that cost usually is around 120 to 130 just for that exact same closure plus commission, plus you have to labor and then you deal with other factors such as having a bad day. A person doesn't have the pep in their voice or they just weren't on that day, or just lack of motivation or called out sick. So with conversational AI, all those things come out and the beauty of it is we're not really talking about a robotic sound. We're not talking about a not conversational sound. We're getting the, we're getting customers what they really want. And in the end of the day, that's really what they want to have happen, even if they realize they're speaking with a conversational AI, they want that. And so I remember you telling me a story we talked earlier about you're on the phone with, I think it's AT& T for four or five hours, going between one agent to the next agent to the next agent, and nobody knew what each person was saying. We can all relate to that, but what if you were to call and you get the conversational AI. Just say you, you just somehow know it's fake or you ask if,"Hey, is this really a really human?" And they say,"No, it's, I'm a assistant. I'm an AI assistant. I'm here to help you." But you get your information within 10 minutes. Would you really care? So that's really where I believe this is all going is most of the time. You can't even tell, but, with these results and these use cases, it's literally mind blowing. And that's why we're excited to be part of that.Isar Meitis:
This is amazing. And I'll summarize some of the things you said, and then I'll add two cents of my own. So first thing we're saying the capacity is basically endless as long as you have. Either the leads or the contacts, right? So it doesn't matter whether it's customer service or sales. If you have the people you need to contact, then there is no capacity limit. It's just, okay, let's add another server in the cloud and this thing will run a hundred thousand calls a day. If we want it to, if you have enough people to contact. So that's one thing. The results could be the same, or like I said, could be better and this is the two cents that I want to add. One of the things I was doing in my career, I was running a large travel company and it had a sales call center. And in the cell call center, there's a huge variance of really great salespeople, average salespeople, and below average salespeople, like in any other job on the planet, because we're humans. And what you do day in and day out, week over week, month over month is you find the best calls, you turn them into training materials and then you do training sessions with the people that are below average and the people that are average trying to get them to the level of your best salespeople and it works with some people. It doesn't work with other people. With AI, it's a given. The system will learn how to close more business. The system will learn what is the best phrases to use to deflect an objection, whether in customer service or in sales and so on. So it 100 percent of your agents will keep on getting better and better over time. So when James is saying, yes, even if I assume the results will be the same, and I think there will be better, my gut feeling tells me they are guaranteed to be better over time. Maybe not better than your best salespeople, but definitely better than your average. And your call center runs on its average, like you have a few superstars and then what you do, you make them the next level, the next domino, right? If this thing doesn't work, okay, let me transfer you to an account executive. Let me transfer you to a supervisor. Let me transfer you to whatever. And then an actual human takes over. That is a rock star that knows how to deal with the extreme situations or with a bigger deals or where. What, whatever the case may be, but the vast majority of cases of both sales and customer service can be handled way above the current average. And like you said, it's maybe the most frustrating thing we have in the world today with large corporations is dealing with customer service, because I have so many departments, you think they're well connected, they never are. You have to tell the story 20 times again and again, and that's if you can get to a resolution in the end. Yeah. And these systems once, and I'm not saying we're there, but once they will be connected to all the data points and all the decision capabilities of, yes, I can do everything in your account and I could do it in five minutes because I have all the access and all the resources and all the knowledge, there's not going to be a customer service call longer than five minutes. And then everybody wins other than maybe people who have spent their entire career being customer service agents which thank you, will probably need to look for another job. So I think we framed the potential outcome very well. And I, if I'm listening to this as a listener, I'm like, okay, how do I get. From where I am today, where I have agents or I'm using a third party, or I'm using an offshore call center, it doesn't matter where you have those resources. How do I go from that to having what you just showed me? What are the steps? What do I need to do in order to like a flow chart? Do this, this, this, this, and that. This is how we take a client from old school to this AI era of customer serviceJames Lindsay:
and sales. Yeah, and that's very insightful to what you just mentioned earlier. I appreciate you adding that to, so the actual flow, very simply usually goes from, if you have a client like our clients who are early adopters, just jump headfirst in, they get it, they understand it and they're, they say, okay, let's go through the process. in the process on that case is really to go through and understand what their current existing sales flow works, how it works. is it working? Is it broken? Do they want to make improvements? Usually there, there's some improvements need to be made there, but they have a working model, so you don't want to break that. You want to improve that. Establish the working model. Once you have the working model of what's there, then from that standpoint, then go through and really understand how their existing equipment that they have functions, how it works. that helps you understand how integrations can go and that this is for a larger client, midsize client who has a call center, who has a decent amount of agents. and then from there you go through and basically let them know, okay, here's the new workflow that we can work with. It can work in conjunction. Let's do a 90 day sprint. Let's take something that's low risk, high return, and let's use that as the individual 90 day sprint. That way they don't have to abandon what they currently have. They can keep what they currently have and run this alongside. So whether we usually do that, whether you're, you have an early adopter who wants who is just dabbling. So if you have someone who's okay, I like this idea, let's test it. And it's say it's a mid range company and they're not early adopters and they may be a little scared of it. Then we still do the 90 day sprint. We just build that small prototype next to it. Feed it in there. Some leads that are working and they could even be secondary ones. you always have those calls where people can't quite get to him. and for whatever reason, feed those into it. And let's see what this thing could do. So that's usually how we move forward with it. And there's usually tweaking timeframes for this can be anywhere from honestly, a few days to, it could be some weeks. Usually it's weeks to implement this. And then some are going to take months because you have to really, you can build it really quickly. You can put it in place very quickly, but it's the iterations. it's the, really the massaging of the, of how you want this to work, because that part is the part that really takes on the dynamic human nature, because then you're getting into the true sales process of what's my star agent look like how can we transfer that into here? there's various things and that's where when you start massaging it That's where you start 10 xing 100 xing what this can do because that's when you're getting into closing sales Which is the whole reason you did this toIsar Meitis:
begin with amazing A few follow up questions. Follow up question number one is as a manager in a company that hears this now, what is the data that I need in order to enable this thing, right? Because it doesn't, it's not a plug and play. Oh, take this thing, click go. And now you have this because it needs to understand, the products that you're selling, the product that you're servicing, the protocols, the sales processes, the potentially learning from previous calls. I don't know what the process, but I think it would be very interesting for everybody to understand what do the solution requires in order to be set up?James Lindsay:
Yes. That's a very good point. So it's actually broken into two categories. You break it into the sales category. So the sales flow, the followup flow, the scripting flow, and then you have the tech portion of it. that's really where the two to divide, but you have to expect you're going to be dealing with both. which you are in a normal call center or in a business anyways. So from a sales aspect, let's talk about that. Cause I think most persons can identify with that. So it basically has in, inside the actual brain of the operation, you have multiple agents. Agents can take on different personalities, different personas. They can be programmed basically with different scripts. They can have different rebuttals in them. So you can choose the same agent. the underlying factor below it all is that the technological mindset of what you want, how your company's brand needs to be presented, that's really what the underlying culture needs to be put in place first. How has that done? That's done by taking recordings of existing phone calls you have of your superstars. Let's take 100, let's take 10, let's take 50 of our winning sales calls that we had that went and maybe not like a dream, but we closed them or some went like a dream and we got them or we set this appointment or we found a very successful way to make this particular customer service issue go be solved. And so when you plug those in recordings, transcripts, and then on top of it, most good salespeople or agents, they have a helper that's there, a cheat sheet of, rebuttals. Here's what I say. So the days of that are honestly gone. But yes, that's how it's trained. And then you can get even more sophisticated if from a technology side, let's talk about that now, unless there's any more questions on the sales. Oh,Isar Meitis:
I do have a question. And I think you touched on it, but I'm trying to make sure that I got it correct. So the agent comes with, let's call it sales skills out of the box. He knows how to run a sales process. He knows how to talk. It knows how to try to get to close a deal. What it needs to is the product or service that this company is selling. And then how Is it best sold in order to now adapt its, let's call it natural sales skills to the environment it needs to operate in, is that a good way toJames Lindsay:
define it? It's a very good way to define it. Yes. Because you never know how that person is coming in. So a standard operation, you may have someone fill out a form that form then goes into a CRM that then gets sent to as a lead that leads then followed up on. So it could have the same flow, when that outbound calls going out through conversational AI, it has a certain temperament, certain personality, a certain understanding when you're talking to that person, because it's more of a cold outreach to them. if you have someone who's an existing customer, you may welcome them with their voice. Oh, Mary, I see that you're calling. Are you calling from this number? We get that with, different services now, so it can do the same thing. So each, depending upon the flow and where it comes from, each individual agent has their own understanding, temperament and the ability to adapt to the culture and the sales or customer service or appointment settingIsar Meitis:
capability. Phenomenal. So I think this leads us straight to the tech side of it, because you said this is obviously connected to the CRM and potentially ERP if it's that kind of company and then whatever other systems in order to, know all these things, right? When a caller calls in and I know their number, it's because it's connected to my CRM and I know their caller ID and I know this is their account and this is what they're probably calling about because there's an open ticket in my customer service ticketing system, right? There's all these different things that a live agent has in front of him, either fully automated or virtual. As we all experience, let me take 50 parameters from you to verify to you and then let me log into your account and then it will take five minutes for me to know who you are. And now you can start telling me what you need. So how does that work? How does all this thing connects to all these different systems? And is it Again, already pre made for most systems out there like Zapier, or is it a custom integration every single time?James Lindsay:
Yeah. So there, it's, there, there's basically two, two versions. the first version is it comes, it already comes basically with its own built in connectors and connections. and with those, it can connect to API. It can connect to Zapier, or some other, third party workflow, that connects directly to the CRM. And then from there, it would communicate back and forth. The current model that we're using that seems to be the best is to use the actual conversational AI as the hub because the technology is growing so quickly and so rapidly that we wanted to stay contained in there and then only spit out information to us through the API through the webhook through a connection that way that goes into the CRM to enrich that information. And then use that CRM as a trigger, just like you would with, say, Salesforce to trigger an email, to trigger an SMS, you have to trigger the outbound phone call, or if you choose inbound, it would automatically accept the inbound call. Add that as a tag into, you say, Salesforce and trigger some other actions. So that's the really simple set up when you get into larger enterprises, it's it becomes a little bit different, not more difficult, but just more sophisticated because they usually have multiple systems running that they all want to communicate with. And that seems to be the trouble with larger companies is they have three or four or five systems that aren't connected. And so our goal is when we do go into an enterprise and they're wanting to really work this the proper way, let's don't do an overhaul, but let's just basically connect these together so that it becomes congruent.Isar Meitis:
Fascinating. so again, to summarize the tech side, this knows how to connect to probably all the major platform routes there like Salesforce, HubSpot. Et cetera. And also can obviously be custom connected to other APIs or webhooks to connect to stuff that he doesn't already know how to connect to out of the box. And then what you're saying is the goal is to have most of the work quote unquote done within this new agent world, where then it either just gets triggered or then gets the outcome. To the CRM platform. So the CRM triggers the call, or if the call comes in as inbound, then that will triggers the call. But then the outcome of I close the sale, I resolve the situation, or I need an escalation of some sort goes back to whatever the end system that is already used in that company. So whether it's a task management platform, like Monday or JIRA or Asana or ClickUp or whatever, or a CRM or whatever, you will know how to connect in the same process. That a human agent would in that company, so it doesn't disturb the current processes in the business. Is that correct?James Lindsay:
That's 100 percent correct. Yeah, 100 percent correct. And many may be wondering too, okay, does this mean that my whole system is robotic? You will still have live agents. You'll probably still have somebody who's there who can always be passed off through, which is the beautiful part. Just like when you're talking on a chat bot that you know is a chat bot, maybe controlled by ChatGPT. If you say,"Hey, ask to speak with a human?" It can really dive you right to a human right away to take care of something. If it doesn't know the answer, if persons just don't want to talk with it, whatever will be the case. But the heavy lifting is done through the conversational A on the workflows, just like you mentioned.Isar Meitis:
Amazing. Okay. Okay. I have one more question that is slightly broader than that. And the broader question is the calls are just part of the process, right? The sales process is a lot, and the customer service process is a lot broader because people today expect to be able to engage with the business through multiple channels. So they could open a support ticket on the website, they could fill up a form that they're interested in a service or a product on an app, they could chat with an agent on wherever, they could send an email or like, how does all of this connect together and do these kind of system can fulfill each and every one of those components. Meaning can it send emails back and forth, be aware of them in the chat and then suggest a phone call and connect you to a live agent because it's the appropriate setup.James Lindsay:
Yes. Yeah. so that is, that's extremely possible. It's being done right now, even without conversational AI. so if you have a CRM that acts as the brain, acts as the hub, that really all of that, that data is connected to, then, just like with, other email services, or, if it's a, Klavio or some other service where you can have automatic workflows that automatically go through send this email campaign out here, switch to SMS, switch to some other type of service that they want to do that's completely doable now. So it's not unheard of conversational AI to it. It just becomes part of the ecosystem. it just becomes a big fix for a problem that's out there right now that most businesses recognize.Isar Meitis:
So to clarify, this solution on its own really takes care of the conversational voice side of things, and it will integrate with other solutions and now it's up to each business or an integrator or whatever to come and figure out, okay, how does that integrate with the chat solution that we're going to have? And how does that integrate with the email customer service platform that we're going to have? And I think the future is. From a customer service perspective, highly integrated where all these things talk to the same AI brain that will know how to address them in the most effective way for whatever scenario, type of client, ICP situation, products, age group, like whatever parameters are going to be there, it will know what's the best way to resolve that. And we'll probably address it that way. And I think that's where everythingJames Lindsay:
is going. I agree. I agree a hundred percent,Isar Meitis:
one hundred percent. James, this was brilliant. I'm listen, I've seen this coming. Like I spoke about this several times before in the podcast saying, I think one of the industries that's going to get impact the most and very quickly, like as far as, okay, this is taking big chunks of a particular market and or a particular, Profession, customer service and sales call sales, like sales call center people are going to be significantly impacted. I just did not know it's happening that fast. And what you just showed me proves that it's, and this is now, right? We are at the. The fourth quarter of 2023, when this technology really evolved in the last year and a half, maybe two years, and it's moving stupidly fast. Like I think in 2024 and into 2025, this thing is going to become the only solution people are going to go to because it won't make any sense to hire 200 sales call people or get offshore operations and so on. Like people do today, because everything we said in the beginning. It costs less. It's available 24 seven. It can speak in any language. It gets better over time. It always follows the procedures and the protocols. It never has a bad day. And so on, before I let you go, if people want to find you, follow you, read your content, work with you, what are the best ways toJames Lindsay:
connect with you? Yeah. Thank you. Yeah. LogicLabsAI. com is the best way to go and look around and see what we're about. You can schedule an appointment, a meeting to, to chat. of course, available on LinkedIn. LinkedIn is one of the best ways to, to contact. James, Lindsay, we're there. Also Logic Labs AI is there also. And, those are usually the best ways to get in contact with us. And, there's emails available on the website or, DM us. We'll be happy to answer any questions or just chat or give some additional information aboutIsar Meitis:
it. Thank you. this was awesome, James. I really appreciate you taking the time. I really appreciate you sharing, this with me and the audience. I really think what you're doing is at the, forefront of where AI is going and I think just to finish up on this. If you broader the view for a second beyond business and customer service and so on and thinking where this is going as far as, Oh, I can have a conversation, a deep conversation on a topic, any topic on somebody who is an expert on that topic that understands me and my needs. And do I want that for other things like mental health or assistance in cooking, or like literally anything that you want to talk to somebody that can walk you through a process or help you solve a problem. This is where we're going. And whether you're happy about it and you think it's good or bad or it's going to destroy humanity or it's the best thing that we can wish for because people who need support can get support right now that will understand them very well. Whether you agree with that or not, whether you like it or not, this is where it's going. And I think as a society. We need to figure out what that means across multiple aspects. And I'm sure you have a few words to say about what I said. and then I think we can say goodbye because I think we touched on a lot of things.James Lindsay:
Yes. Yes. that's a, it's a beautiful point you brought up because, I really didn't think about that completely until you just mentioned it. But thoughts that were going through my head was, those things that you just mentioned, whether it's mental health or talking to a doctor, or an attorney for that matter. those are all things that either people could be too shy to talk about, or they truly need an emotional touch. And that emotional touch doesn't happen always. So there's a lot of SMS chats that are there. there's chatbots that can give you some good advice or good things. But I think when you have that voice, That's there with the emotional portion of it. The emotional touch that happens. I think there's a deeper connection that happens, especially when you start to speak to people in their native tongue. It would really start to reach their hearts more than just seeing words on the screen.Isar Meitis:
Agreed. James, again, thank you so much.James Lindsay:
Yeah. Yes, sir. Thank you.Isar Meitis:
What a fascinating conversation with James. We touched on a lot of critical things, And the two most important things to take as far as I'm concerned from this conversation, one is the technology today can already do the job of call center people in most cases. And the other is that this is the worst AI we're ever going to have, meaning this is going to keep on getting better and better and I've already listened to conversations that James shared with me of the previous version. The current version, and these are already night and day. This is probably going to replace most call centers on the planet in two to three years, and similarly replace most of the conversations with clients in general, which actually means that the most important thing you can do as somebody who has a contact with clients, whether you're a salesperson, customer service, et cetera, is to develop the best human relationships you can, because this is what's going to differentiate you from the competition. Because if everybody's using AI to deliver customer service and do sales calls and follow up and so on, who's going to win. The person that has the best personal relationships with the person on the other side. And now to the big piece of news I promised you in the beginning of the conversation. And it's not even totally news yet because it wasn't formally announced, but it was leaked by a few OpenAI customers. So OpenAI, the company behind ChatGPT has released to some of its users a true multi modal AI model. And what does that mean? It means, and I'm quoting now, you can work with PDF data files or any document you want to analyze. Just upload and start asking questions. You have access to browsing advanced data analysis, and Dall-e in a single solution, meaning you don't have to jump back and forth between them. So what the hell does that mean? For those of you who don't know, ChatGPT added more and more features in the past few months, like the ability to upload files into what they call advanced data analysis. And you can ask it question and analyze really interesting information, whether you upload a PDF or an Excel file or a Word document, you can ask questions about all these components combined. But in there, you couldn't upload images because it did not know how to analyze that, but it did know how to analyze images in the main regular chat. That being said, you cannot create images as of now in the regular chat, you have to select the Dall-e module. Well, all these silos are going to turn into one system, which means we will be able to do magical processes within one spot. As an example, you'll be able to upload multiple touch points with potential clients. Information you have about them from third party platforms. Emails that you've exchanged with them. Transcriptions of conversations that you had with them, et cetera, to build a model of that client, have it, analyze it, create pricing models specifically for them based on the requirements that they've raised in the conversations with you, adding additional information like your financial costs and so on, creating a proposal, adding that proposal into an email that it can write for you and even add graphics for the proposal. You can take the same thing for HR use cases, such as creating training for the company. You can analyze the results of the internal testing. And benchmark that you're doing using advanced data analysis. Then you can use that information in order to define the knowledge gaps. Then you create the actual lesson plans and PowerPoint presentations, including the graphics that goes into them using Dall-e. And then you can package it all together and put it in whatever software that you're using with about 20 percent of the work that it would have required doing it the old school way and the same thing is true for marketing. And the same thing is true for finance. And the same thing is true for every aspect of the business. This is a true game changer. And if you haven't used each and every one of the tools individually, I encourage you to do that right now. Follow people who do this. Obviously you can follow me on LinkedIn. I share different examples on how to do that, but there's a lot of other people you can follow and start experimenting yourself because once all these tools are together, it is going to revolutionize every aspect of our businesses. I'm going to end with that today without sharing additional news. That being said, I'm going to share links to additional news on the show notes so you can go and check them out if you're interested to see what happened this past week. And until next time, have an amazing week.