Latest / Speak In Flow / Why Most AI Projects Fail

Transcript

Why Most AI Projects Fail · Speak In Flow

• 0:02
  Welcome, dear listeners, to the Speak and Flow Podcast, where we dive into unique strategies and stories
• 0:08
  To help you and your team achieve maximum potential and momentum through communication.
• 0:15
  Today, I have an amazing leader. He is in the AI space. We're going to be talking about the conversations that are happening with AI. What are the things that we're resistant about, and what are…
• 0:27
  Um, the conversations that we need to have as leaders when it comes to AI.
• 0:32
  And he is the CEO of DIFCO.
• 0:35
  He is an advisor to
• 0:38
  Alchemist Accelerator. He's an expert speaker and leader in the AI space in cybersecurity and also product development.
• 0:47
  As an advisor, he likes to help companies scale through AI, and he's
• 0:52
  Also, passionate about aviation. Uh, he loves coffee.
• 0:57
  And… and so I welcome Vadim Peskov.
• 1:02
  Hi, Vadim.
• 1:03
  are great to be here, and thanks for inviting. It's, uh, definitely will be a fun conversation.
• 1:08
  How AIA will change the world?
• 1:10
  how AI will change the world, as long as we don't kill ourselves.
• 1:14
  Yeah, that's a second chapter. We'll… we may skip it. We'll see.
• 1:16
  I love it, I love it. So tell me, Vadim, what are you excited about when it comes to DivCo?
• 1:24
  Your company.
• 1:25
  Yeah, so, just for the context, again, we are providing software development services for last
• 1:31
  17 years, and uh…
• 1:33
  We're doing a lot of work with AI, specifically, eh?
• 1:36
  Um, and we see a lot of different startups and a lot of enterprises trying to
• 1:41
  change how they work, and uh… how they do things, basically, with use of
• 1:47
  or differently and different ML models.
• 1:50
  And I think this is the really interesting, uh, edge to be on, because we…
• 1:55
  see how things is progressing and changing in direction.
• 1:59
  of what the future will look like in…
• 2:02
  like, half a year, a year, two years, etc.
• 2:04
  So, I think what I'm really passionate about…
• 2:07
  that the use cases and how we use the current software
• 2:11
  And what we can get from the… our current AI models.
• 2:15
  As well as, uh, the current…
• 2:17
  products that we use day in, day-to-day life.
• 2:20
  will be actually much more advanced and much more interesting
• 2:24
  in near future. So, I think it's a really interesting, uh, way to see how we progress in this society.
• 2:31
  And the acceleration that we see right now,
• 2:34
  It's absolutely crazy.
• 2:36
  I don't think we ever saw so many companies growing so freaking fast.
• 2:42
  And I think we will have even more crazier stories.
• 2:46
  Yeah.
• 2:47
  So, this is what we're doing, and this is what is keeping me motivated for this.
• 2:52
  How long have you been in the AI space?
• 2:55
  That's a good question. Uh, right now, it's… I think it's 8, 9 years when we started…
• 3:01
  initially are doing a lot of, uh, mostly as well, computer vision and ML work.
• 3:05
  Back then, um, and a lot of those things is called the big data, et cetera, it's a big…
• 3:12
  the different kind of naming for the same things, in a way, um…
• 3:15
  Ah, I see.
• 3:16
  Uh, however, it'd been a long time.
• 3:20
  That's so… because from an outsider who had just been exposed to AI this year,
• 3:26
  And now I see so many people that have been in the AI space for many, many years, like, but they're… all of you have different
• 3:33
  expertise in the AI world.
• 3:35
  We've been cooking this for some time.
• 3:36
  But still, you've been cooking this! It's been brewing! That's so fascinating! And so, thank you so much for joining the conversation today, and…
• 3:45
  And so, again, you see how, um, how AI can really change the way that we grow our businesses in such a rapid pace.
• 3:53
  And, um, what are you hearing with the conversations of the people that are just resistant to it? What are they saying to you?
• 4:02
  To be honest, I don't have a lot of conversation about resistance.
• 4:06
  Okay, good.
• 4:07
  I do have a lot of conversation of misunderstanding.
• 4:10
  Okay.
• 4:11
  Um, because in many cases, people…
• 4:14
  uh, tried some tools.
• 4:16
  And try to see what is possible,
• 4:20
  And they wasn't able to get the result.
• 4:23
  One way or another, and in many cases, they just use the tool in their own way.
• 4:29
  So basically, in many cases, they will take a…
• 4:33
  large tool, and we'll try to…
• 4:35
  ask it even larger questions sometimes, and do not provide any context.
• 4:41
  Um, and…
• 4:43
  like, hey, I'm asking ChatGPT for something,
• 4:46
  Um, and it's provided me a terrible answer.
• 4:49
  is great. Um…
• 4:51
  Uh, and I'm thinking that it's a useless tool. The reality is…
• 4:55
  In many cases, you need to explain what the heck you want as a result, how you want to process the data, how you want the things…
• 5:02
  Again, in plain English, no, no…
• 5:04
  again, magic here.
• 5:06
  The problem is that many people…
• 5:08
  Um, try to kind of, like, do a prompt with one line or two lines is great.
• 5:15
  It doesn't explain you what to do.
• 5:17
  So, if you would ask your coworker that's never worked on this problem before, you maybe will spend 5 minutes.
• 5:24
  Maybe more? Explain what exactly you want this person to do.
• 5:28
  And after this, you can make this work.
• 5:31
  Mm-hmm.
• 5:32
  In some cases, we'll have a prompt that
• 5:34
  more than a page long to actually execute on things that we need.
• 5:38
  Because, again, you didn't want that some…
• 5:41
  A random result, you want to…
• 5:43
  exact result, what you're looking for,
• 5:46
  And sometimes you need to explain the things that need to be done.
• 5:50
  Right, and providing the right context, and you're saying that the…
• 5:55
  Having the right information in there sometimes could take up a whole page versus just a one sentence.
• 6:01
  Yep, absolutely. Again, especially if it's a…
• 6:04
  not like, hey, fix the…
• 6:06
  grammar in this email, but, like,
• 6:08
  Right.
• 6:09
  help me actually research some kind of topic.
• 6:12
  Right.
• 6:13
  Um, so if you ask in AI, help me
• 6:16
  find a better future for myself, my career, like, some career advices.
• 6:21
  I have no freaking idea who you are.
• 6:23
  beside your name, which basically sometimes will get not a lot of information that you can find.
• 6:29
  online about you, or about your experience, about your strengths. So, it is a lot of angles that you need to add.
• 6:35
  to get the result. Same for the business as well.
• 6:38
  And earlier you said that companies are misunderstanding of how to use it.
• 6:45
  Have you run into a company that have used it, um, not knowing how to use it well, and losing a lot of money, or a lot of time, productivity, effort?
• 6:56
  Do you have an example of that?
• 6:59
  I don't have a lot of examples of, like, losing specifically money, I don't think…
• 7:03
  um… I've seen those cases so much.
• 7:06
  But I seen a lot of missed opportunities.
• 7:08
  where, um, you just…
• 7:11
  Having a lot of, uh, guardrails installed.
• 7:14
  Uh, and you cannot do these things and this thing, and you're worried about the data so much.
• 7:20
  Um, that you will share with some model,
• 7:23
  Um, and company not providing you an on-prem solution that actually sharing your data,
• 7:30
  And, again, it's not a complex thing, so it's a, again, um…
• 7:36
  But when we do EA transformation sessions, when we help those companies,
• 7:40
  In many cases, we see that managers and management
• 7:43
  See, Elise is like, yeah, let's actually make it work.
• 7:47
  Um, but they…
• 7:50
  themselves do not have a deep understanding of this. And my strong belief…
• 7:55
  And maybe I'm completely wrong here?
• 7:57
  But my strong belief is you, as a founder, or you as a C-level exec,
• 8:03
  You need to understand what this tool is doing.
• 8:06
  Um, again, you don't need to specifically understand the technicalities of this, you don't need to
• 8:11
  called anything yourself, sure.
• 8:13
  But you need to understand
• 8:15
  what is the capabilities of this? Because, let's say you have a construction and you…
• 8:21
  buying a truck, you want to understand the characteristic of this truck.
• 8:26
  can distract live the certain weight, and can it do X, Y, and Z things that you need?
• 8:31
  Sure, you want to understand this, and you didn't want to buy a different version that will not be able to hold the weight that you need.
• 8:39
  Same story here, but in many cases, people…
• 8:42
  Just do not spend enough attention there, and they believe that, hey, it's a…
• 8:47
  I played with ChatGPT, or I played with Clod, or I played with Gemini,
• 8:51
  Um, and it's good enough for me to understand how this works.
• 8:55
  The truth is, probably not.
• 8:57
  Got it.
• 8:58
  Probably you need to spend a little bit more time understanding that gentica approaches, you can understand
• 9:03
  how you can actually apply to your business, what the actual limitations are,
• 9:07
  what functions you can replace, what…
• 9:09
  functions you can enhance with AI.
• 9:11
  This is interesting. So, as an executive, you're finding that they're not as resistant as before, they're more misunderstanding.
• 9:22
  They don't really understand how to use it, the power of it, and how they can even apply it to a business.
• 9:30
  Yep.
• 9:31
  to be strategic, right? It's more about, like you said, they just using it to type in some prompts in Claude, or in ChatGPT, and they think that that's
• 9:41
  enough in understanding.
• 9:43
  Yeah, so, um, and again,
• 9:45
  different companies, different execs, different…
• 9:47
  Yeah, right.
• 9:48
  levels. But that's why we…
• 9:50
  again, for the last couple of years, we, like, really, uh…
• 9:54
  love to use a founder mode, uh, uh, examples.
• 9:58
  Uh, of many, many, uh, companies here, and we have a lot of jokes about what the founder more is and why it's a legal acceptance.
• 10:05
  Okay.
• 10:06
  But in reality,
• 10:07
  you need to understand what the heck is going on.
• 10:10
  Yes.
• 10:11
  Uh, because if you don't…
• 10:13
  Um, diving to…
• 10:16
  the actual opportunities of what your business can do with this,
• 10:20
  Yes.
• 10:21
  you will be replaced. And…
• 10:23
  What I see, like, in…
• 10:24
  The CEO will be replaced?
• 10:27
  For the company, the organization.
• 10:28
  the businesses will die. The whole company, the whole company, because…
• 10:30
  Yeah, correct, correct.
• 10:32
  In many cases, I see that, for example, let's take the insurance company's example.
• 10:39
  I strongly believe that it's possible to build a large insurance company with, uh,
• 10:45
  maybe hundreds of people, not thousands or tens of thousands, that'd be.
• 10:50
  heavily now, and do the same work even better.
• 10:53
  Mm-hmm.
• 10:54
  Which means, uh, much lower capex.
• 10:57
  Which means much better profitability, which means we can provide
• 11:01
  better quality of the service, and charge less.
• 11:05
  And I don't think it's a problem to build those type of businesses. I think it's easy.
• 11:11
  enough, so many, uh…
• 11:14
  people will have those kind of examples, build and really replacing.
• 11:19
  So, last year we saw… sorry, this year, we saw the company that was acquired for $80 million with 1%
• 11:26
  Uh, that built this company?
• 11:29
  Which is, hey, one person was…
• 11:32
  80 mil, uh, for, like, a couple years' worth of work.
• 11:36
  It's not really a sophisticated product, but, I mean, it's a good product. The point is,
• 11:42
  had no one else. It's just one person, specifically, that
• 11:46
  did everything there. There's not 10 people, not 20,
• 11:51
  1. And again, it is a wonderful journey, and I think this is what is possible right now. And I think…
• 11:57
  we may have a billion dollar re-exit
• 12:00
  with the folks that maybe one or two people, etc. I still believe if he had maybe 10 people, it would be a billion bucks.
• 12:09
  not, uh, specifically, uh, 80 million, but…
• 12:13
  I think it's still a great opportunity.
• 12:15
  for the growth. So you have a lot of…
• 12:18
  ways to see this, and…
• 12:20
  Eve the one guy can do this,
• 12:23
  is most likely you can actually…
• 12:26
  change the logic of the current business in the same way?
• 12:30
  that much more dramatic, but people scared because they do not understand.
• 12:35
  And they typically don't spend enough time understanding.
• 12:38
  And they think it's toys.
• 12:40
  While people on the other side actually earning money.
• 12:43
  Yeah, so what advice would you give an entrepreneur, you know, with
• 12:49
  some employees, like 200, 300, 400 employees, how can they start to understand
• 12:57
  how AI could support this business.
• 13:01
  What advice?
• 13:02
  I think it's a… it depends on the organization, to be honest.
• 13:05
  Because if you have a…
• 13:07
  rights to do things.
• 13:09
  in this organization. Um, I think it's a wonderful opportunity for you to start implementing and learning those things.
• 13:17
  And I'm not saying that it's really, uh…
• 13:21
  requiring a lot of technical effort.
• 13:24
  But it is a basic, uh…
• 13:27
  visual programming that you can do with some tools, etc.
• 13:31
  Sometimes there's maybe some code, but to be honest, AI can write the code for you, so you don't even need to program.
• 13:37
  Uh, yourself, so I think in many cases, it's possible to do a lot of things.
• 13:42
  that, um, many people just, hey, I'm not a programmer.
• 13:46
  I was like, sure, but you don't need to be.
• 13:49
  It's maybe need to be executed in a code in some cases, or it's better executed with a code in some cases.
• 13:56
  But I think we will have a lot of tools, and we already have them, that, like, require zero coding.
• 14:03
  that you can build a great prototype, so you can…
• 14:06
  create a lot of internal tools.
• 14:08
  that will be used, um…
• 14:10
  Uh, for different purposes, etc.
• 14:13
  Uh, that can be, uh, much better than a current.
• 14:17
  flows. Sure,
• 14:18
  So how do I know a flow, and what even to program, right? There's probably so many workflows, so many different…
• 14:25
  tools, like, how would I, as a CEO,
• 14:28
  looking to grow and scale my company, how do I know?
• 14:32
  If you don't know your…
• 14:35
  flows of your processes in your organization, you should be replaced as CEO.
• 14:39
  I don't know, meaning, like, I do know, but how do I know which ones to start with?
• 14:44
  Sorry.
• 14:45
  Oh, um, which one is taking the, uh…
• 14:48
  all, uh, basically, the more team effort.
• 14:52
  Okay.
• 14:53
  To complete. Um, and my point is,
• 14:56
  In many cases, just to be clear,
• 14:58
  The right answer may be not in AI.
• 15:01
  Wright answered maybe in a different tool. I'm not saying that AI is a solution for everything.
• 15:06
  No, it's not.
• 15:07
  It's not? I thought it is!
• 15:08
  No, it's not.
• 15:10
  It's not so accurate in some cases, and it will have some issues.
• 15:15
  And let me kind of maybe, um…
• 15:19
  draw the picture of the things, how I think the future interfaces will look like.
• 15:24
  Okay.
• 15:25
  Uh, in many tools that we used in a day-to-day basis.
• 15:27
  I think we will not have…
• 15:30
  People managing the tools?
• 15:32
  Specifically, and hey, we need to create this report, we need to do this, we need to send this to Joe.
• 15:38
  I don't need to review this, etc. So, I think we need to see the interfaces that we will have,
• 15:45
  is, uh, the tasks that prepared for us.
• 15:49
  already done, the report was already generated, the result was already there,
• 15:54
  You just need to review this and click send.
• 15:58
  or approval, or whatever.
• 16:00
  And in this case, you're going through the tasks,
• 16:04
  that maybe you will take maybe 15 minutes, maybe 30 minutes of your day going through those things, versus spending
• 16:11
  a half a day on doing those tasks.
• 16:14
  So you're reviewing the results. So, my point is,
• 16:18
  You can ask here to, again, fully execute this flow,
• 16:21
  Or you can ask, or be much more careful and execute this in a way that humans still approve this.
• 16:28
  So, human review and understand what is going on, and saying, yeah, here's a mistake.
• 16:34
  We need to fix this, we don't want to disclose this, we want to disclose this.
• 16:38
  Etc. So, it is a lot of…
• 16:40
  details that you can, uh, add to the flow.
• 16:44
  that will be against steel,
• 16:46
  Uh, involving a human in a loop.
• 16:48
  Um, at least in the beginning.
• 16:51
  Because, again, strongly believe that in a lot of things in finance and in a lot of things in accounting could be really easily automated.
• 16:58
  Uh, again, uh, maybe same for, uh, definitely same for marketing, and definitely a lot of things in terms of the management.
• 17:07
  Again, uh, why you would ask the same thing…
• 17:13
  every day of the…
• 17:14
  some team members. Like, hey,
• 17:16
  Can you send me this report?
• 17:18
  Yeah, I can do this for you, and can come to you and say, hey,
• 17:22
  Here's the report, here's the key things that I saw in the report that
• 17:27
  Um, not right. So, I asked the question, here's the answers to the questions.
• 17:33
  So, again, not complicated at all.
• 17:36
  again, different businesses, different flows.
• 17:39
  But, again, speaking of example of insurance,
• 17:42
  Again, we can profile, we can do a lot of research and understanding with that human
• 17:48
  Um, I'm underwriter definitely will not have
• 17:51
  time to do this. But we can learn from the human
• 17:55
  Uh, to understand what the things that we need to look for.
• 17:59
  to be able to, again, decline an approval wherever we need to, um, estimate there.
• 18:04
  Right. Right. It sounds like a lot of the functions of a business will still be there, but it might, rather than have 10 people, maybe 1 person.
• 18:16
  Yep.
• 18:17
  Does it? In marketing, or maybe, like, it could be completely wiped out, but maybe you just have one marketing person overseeing all these things and just saying approved or not approved, or edit.
• 18:25
  And same thing with finance, same thing with accounting.
• 18:29
  Yeah.
• 18:30
  You're absolutely right. So, I don't think it's a…
• 18:32
  the case is that you would want to replace the humans with AI. In some cases, it's true, but…
• 18:39
  Yeah.
• 18:44
  Great.
• 18:45
  In many cases, it's actually different. You want to enhance humans with AI. You want to give them a tool to be more productive.
• 18:48
  Right.
• 18:49
  So you can have the same workforce,
• 18:50
  Right.
• 18:51
  but just do more things?
• 18:53
  with the same, uh, things. Again, coming back to our insurance company. We can approve
• 18:59
  10 times, 50 times more, uh…
• 19:02
  uh… estimates.
• 19:05
  really, really quickly, just because…
• 19:07
  He already did the job, and humans maybe just need to approve this. Or maybe it's…
• 19:12
  If it's meeting some level of thresholds, it could be approved automatically, and we don't really…
• 19:17
  to have the approval.
• 19:20
  are of the human.
• 19:22
  So, again,
• 19:24
  I think we can see the functionality that's done differently,
• 19:29
  Uh, rather than trying to again.
• 19:31
  accomplish this using, uh, the traditional methods.
• 19:35
  I'm excited. I'm… I know that how it's impacted me and impacted my business, and I could see that at a larger scale.
• 19:42
  what you're talking about.
• 19:45
  And how it supports. And so, can you share, um, a client success? You don't have to share names. I know you might be probably under NDA, but
• 19:53
  what, um, client success have you seen? Results?
• 19:58
  From your work.
• 20:00
  So, let me kind of give you maybe a couple of quick examples of that I think is
• 20:06
  making sense in this case. Because, uh, like, we can take…
• 20:11
  Uh…
• 20:13
  regular flows that exist, for example, for…
• 20:16
  uh, really boring…
• 20:19
  thing, like, again…
• 20:23
  accounting, or, again, um, financial planning, etc.
• 20:26
  Um…
• 20:28
  Some people love it.
• 20:29
  Some people are loving, but you and I are in agreement, I don't…
• 20:30
  Again, I'm a big fan of…
• 20:32
  financial portions, I will probably die if I will be an accountant.
• 20:36
  Yes.
• 20:37
  Um, again, I have a lot of respect to those people, because I don't understand how… my brain just wires differently.
• 20:44
  Uh, I were… again, we see… so we built into the system right now that, uh, utilizing AI to
• 20:51
  do this, uh, functionality.
• 20:53
  And, oh, funny enough,
• 20:55
  And that we built in this system for public sector.
• 20:59
  And the core problem that we have…
• 21:02
  We need to convince the client, which is a state's, uh, that we'll be using this,
• 21:07
  Uh, we need to convince them to actually, um, use this tool in a way that will allow them to
• 21:14
  uh… have…
• 21:16
  probably 10% of the stuff that they currently have.
• 21:19
  Um, because…
• 21:21
  a lot of things is actually automated.
• 21:24
  Um, and they're excited about this.
• 21:26
  The unions does not.
• 21:29
  But the fun thing is…
• 21:31
  This is the logic that, uh, we can bring to the table. This is the…
• 21:36
  kind of approach that could be much different.
• 21:40
  And, again, same with multiple other things that we build in, um, in day-to-day life.
• 21:47
  Uh, where we can basically change the flows and make it…
• 21:52
  that, again, you collect information with AI much faster.
• 21:57
  process different kind of information, or you provide the different insights to the end clients, or…
• 22:05
  to the team. I think, like, um, in…
• 22:08
  I can bring you one example of the…
• 22:10
  one enterprise client that we have, uh…
• 22:15
  First, uh, company that's dealing in manufacturing and chemical industry.
• 22:19
  So, we have a tool that used to create different types of reports.
• 22:25
  Oh, there. And it is a large…
• 22:28
  system that we pull in the data from, and we generate reports.
• 22:32
  Um, previously it was a team that did this, currently it's done by EI, uh, and we have a human that's reviewing this.
• 22:41
  For legal purposes. However, it's all done with AI, with basically minimal intervention
• 22:47
  from the human. And, uh, another thing that's important, uh, we have interface for the inclines not just to
• 22:55
  review the PDF reports, which they had.
• 22:58
  Now it's a tool where they can actually ask questions.
• 23:03
  And they can understand what is this going on, etc.
• 23:07
  And we saw being examples when they share this, not with the technical folks, they share this with the VPs and C-levels.
• 23:14
  that can ask a question like, hey,
• 23:17
  What about this, how this is progressing, how we can optimize the results, etc.? Can we get more out of this turbine, etc.?
• 23:25
  And the answer is most likely yes, and here's what you can do, etc.
• 23:29
  Again, it's have all the disclaimers that you need to have a technical review, etc.
• 23:34
  Sure. It's really, again, critical manufacturer, we cannot just, like…
• 23:39
  We can assume things. Uh, we need to have a multiple confirmation. However, based on data,
• 23:44
  we can make this tool that will tell the client multiple options how they can
• 23:50
  improve this, and how they can make it really, really quick.
• 23:53
  on a board meeting versus, hey, can we send this
• 23:56
  someone and get their response in one month. No, they can do this in, like, 15 seconds.
• 24:03
  Wow. Wow.
• 24:05
  That's amazing. That is amazing.
• 24:09
  Congratulations on that! Congratulations! You're at this really inflection point with the work that you do.
• 24:16
  And really seeing the results for your clients.
• 24:19
  I think we'll have much more, uh, in this regard.
• 24:21
  Yeah.
• 24:23
  that, uh, will be much more disrupted.
• 24:25
  to the current flows.
• 24:28
  Yeah.
• 24:29
  I strongly believe that if you muster
• 24:31
  understanding of the current tool, you can do much more.
• 24:35
  with this. Because in many cases, um,
• 24:39
  Again, if you…
• 24:42
  I understand many people don't really be big fans of reading manuals, but in some cases, you actually need to.
• 24:49
  Uh, and you can, uh… again, in real life, you can break things and have to read the manual, sure.
• 24:54
  for home appliances, sure, please do this.
• 24:57
  Um, but for everything else,
• 25:00
  you need to understand the capabilities,
• 25:03
  And the point is, those capabilities changing every day.
• 25:06
  Yeah, yeah.
• 25:07
  Each day, I'm reading something in the news that…
• 25:10
  I was telling a year ago that, hey, it's probably 5 years ago away, it will not be possible, etc.
• 25:16
  Yeah, yeah, yeah, right.
• 25:19
  Every day.
• 25:21
  This is absolutely astonishing, and I don't think we…
• 25:25
  Um, actually seen anything like this.
• 25:28
  Right.
• 25:29
  I don't think we'll have a GI soon. I don't think…
• 25:31
  we will… we can ignore the progress that's currently happening with those tools.
• 25:37
  And think, like, hey, yeah, we probably will be okay.
• 25:40
  No, we will not. A lot of…
• 25:42
  Uh, at least, uh…
• 25:45
  Um, thinking, uh…
• 25:47
  type of job, so basically anything that knowledge workers do.
• 25:51
  I think will be replaced.
• 25:54
  Really soon. One way or another…
• 25:56
  or enhanced, if you're smart.
• 25:59
  or replaced if you're not. So…
• 26:01
  Yeah.
• 26:02
  Play… how you want it.
• 26:05
  I love it. I agree. I agree. So, Vadim, so what would you say to these people, these entrepreneurs right now?
• 26:16
  Uh, what is the mantra that you would like to share?
• 26:21
  I think the core thing is…
• 26:23
  All they're the same.
• 26:25
  It's the same playbook, experiment more, the people that win in business, not the people that have the most genius ideas, in many cases.
• 26:34
  that people that, uh, dumb enough to try this multiple times?
• 26:38
  Yeah.
• 26:39
  Uh, and I think…
• 26:41
  the core point is trying more,
• 26:44
  and experimenting more, and seeing opportunities, and being bold.
• 26:48
  own solutions.
• 26:50
  Yeah.
• 26:51
  Because someone will.
• 26:52
  And someone will… will take your share of the market.
• 26:56
  If you will be slow, and if you think that…
• 27:00
  hey, the tool is not there.
• 27:02
  The point is, the tool is not there today.
• 27:05
  you need to think how this tool will enhance in a couple of years, so…
• 27:10
  you will see the opportunity to implement this to your business that will be much more disruptive.
• 27:16
  Because someone actually doing this right now?
• 27:18
  And they're changing the way…
• 27:20
  how they will compete, and you will not be able to outsmart them
• 27:25
  and do this much faster than they did.
• 27:27
  Just because they're already running in a…
• 27:32
  basically speed of sounds, you need to be…
• 27:35
  Right. Right. Like you mentioned, so take it, read the playbook, understand it, implement it, do it fast, do it quickly.
• 27:36
  Even faster than this.
• 27:45
  And I would suggest hire someone as an expert like Vadim, so you don't fall too hard.
• 27:52
  And so how…
• 27:54
  Experiment!
• 27:55
  Or experiment yourself. I'm really encouraging people to experiment themselves.
• 27:56
  experiment themselves, but I'm also inviting them to contact you in case they have questions. Would that be okay?
• 28:04
  Great, and so…
• 28:05
  Yeah, absolutely. Yeah, you can Google me and find me on LinkedIn, or, uh…
• 28:07
  X, etc.
• 28:08
  Wonderful! So, uh, that's how you… you know, get ahold of Adeem, you have any questions about how to implement AI into your company to accelerate, grow, and scale.
• 28:20
  Uh, yeah, contact them on Google or LinkedIn.
• 28:23
  And I… that's it, that's all I have. I really appreciate your time today, Vadim. I think that we had a great conversation.
• 28:29
  Thanks for inviting.
• 28:30
  It was really wonderful to hear. I learned so much. Uh, I am so glad that we have people like you at the helm.
• 28:38
  Leading the charge in the AI space and helping us to grow with it and learn how to use it better.
• 28:44
  Appreciate it, and uh…
• 28:46
  Experiment more, and we'll have fun.
• 28:49
  We'll have fun, we'll have fun! Just don't kill each other, right?
• 28:54
  That's the next chapter. We'll see.
• 28:55
  That's the next chapter! Okay. Alright, thank you, audience, for being here. I'm so glad. Um, until next time, I'm your sister-in-Flo. Remember, anytime you have a chance to communicate, you also have a chance to inspire,
• 29:10
  and make a positive difference in the world.
• 29:13
  Much love, take care. Bye-bye.
• 29:16
  Hi, Vadim.