0:00 Music 0:06 I want to show a use case and constraints where you can use OpenClaw and still have it securely and be sure that it doesn't leak anything sensitive on your site. 0:19 I'll explain why it's set up this way, what are the limitations. 0:23 So going like a little bit in advance. 0:26 Today, we won't give OpenClaw access to any sensitive information like to your email, to your contact data, to your parcels, credit cards, to your personal details or anything. 0:38 But we'll explore several use cases where it can be used securely and understand why and how to prepare it to further integration with other systems. 0:50 If you look around, there have been several experiments, and I've saw some thread on Twitter, now x.com, when someone set up their open claw, gave them access to 100 bucks and said, hey, make me a million. 1:04 And going forward... 1:07 It was like really spectacular to watch how this open floor signed up for different accounts, tried to trade stocks, tried to set up some sales funnels and benefit from like all the general knowledge out there. 1:22 But at the end of the day, $100 spent, nothing made, no profits. 1:27 So it doesn't work like out of the box. 1:31 How I like to explain it, AI amplifies the skills that you have, the talents that you have, and that mission that you bring here kind of to the human civilization. 1:42 So yes, anyone can do the app. 1:45 Anyone can repeat it. 1:47 But can they do it with the same level of attention to detail, the same level of passion and dedication? 1:54 Probably not. 1:56 So if you are on a mission, if you understand well what it is you're after and what it is that you want, 2:05 I think AI can help you to build it up the way that no one else will be able to repeat reliably with the same level of dedication. 2:16 Because at some point of time, the differentiation becomes not about what to build, but about the mission and the values that you deliver. 2:25 So just give your very high level example why I think so. 2:33 You kind of, if you look at the use cases of AI adoption in the industry, like overall, and this is like actually the perfect segue to our first slide in the presentation, 2:48 actually only 6% of the use cases 2:52 proved that they have any positive for a while at all. 2:55 So, like, extrapolate that to the market and the 3:00 product you mentioned built. 3:01 In 6% of the use cases, anyone trying to do the same can probably get some value of it and can make money. 3:09 And anyone can, but do they have that in themselves? 3:16 Do they have that understanding of how to build this together? 3:20 And why this fails to me looks pretty clear and obvious. 3:26 AI is compressed archive of all human knowledge that accumulated over the course of civilization. 3:35 So everything we ever know, everything in public domain, of course. 3:40 Everything that was ever put on the internet was compressed, categorized, systematized, and packed as like one huge archive. 3:49 And we can, now with school, we can ask this archival question, but eventually whatever answers we'll get are just permutations of different patterns matched to your question. 4:02 So you get out of this huge archive, basically some random permutation of previous patterns, some of which are useful. 4:13 And this is also kind of, I can remember my university professor or... 4:22 Was saying this thing about all mathematical models. 4:26 And his statement was, you know what, all mathematical models 4:30 models, extrapolation projections that we're going to build are on all of them. 4:36 None of them actually reflect the actual world. 4:39 But most of them that I will teach you are useful. 4:42 So in this way, whatever products, outcomes that you can build with AI, they are useful. 4:51 But they're always wrong. 4:52 And there will always be kind of this space dimension where you can fit in and find a way to extract positive value from deployment. 5:04 If you follow, again, your passion, your excitement, if anyone shares it, they can probably match you. 5:11 So it's like, for me, a game of people using this incredible technology to amplify and deliver to the world what they want and what they have. 5:23 Right now, everyone has tried chat GPTs, clouds, or any other solutions, co-pilots in their daily workflows. 5:34 And hardly anyone haven't heard about AI because all recent years it's been all around. 5:40 So they're like some hard mess. 5:42 This is just something that I quickly looked up online when preparing materials for this presentation. 5:51 Only 6% of the use cases. 5:54 Actually have and see some value. 5:57 And this comes to like, 6:00 different reasons to that. 6:02 And one goal that I have today is to equip you with the knowledge and understanding, first, why 6:10 it's not so common for everyone using AI to get the benefits and to have positive feedback and have the return of investment on their time and benefits. 6:22 And the other understands why it happens. 6:24 Today, what I want to do is first dive into some fundamentals and give you some very high-level basic understanding of the technology, of its limitations, of where it makes sense to use it and when it doesn't, 6:39 and also how to start using it for yourself responsibly. 6:43 Into some limited constraint and extent. 6:46 So in the second part of this day-to-day, we'll set up open core for like every one of you. 6:53 So again, like repeating myself, we're not going to give it access to anything like sensitive private, and I'll explain kind of why, but we'll explore use cases that will make it useful for you, like in your daily use cases in your daily life. 7:07 So you can't just ask AI, hey, make me some money. 7:10 Unfortunately, it doesn't work. 7:12 Maybe it will in a couple of years, but that will be a different story. 7:19 So in the nutshell, also another thing that is happening, like everyone is trying to use AI extensively. 7:26 And the trend in 2025 compared to the 7:30 previous year is also pretty obvious. 7:32 Only 70% of the projects across the whole industry were abandoned since 2024, but in 2025 and more. 7:42 It is pretty explainable because most of the experimentation becomes really low. 7:48 So it's easy to try things, it's easy to experiment, it's easy to build proof of concept. 7:54 Try your ideas and eventually fail to learn something new. 7:59 So from early failures gives an opportunity to validate your ideas, to try things, understand what works, what not, not to prepare yourself for building something that actually works and you can operationalize it. 8:14 That's also kind of another part today. 8:16 We'll kind of get these fundamentals to see how you not only can experiment, but how to prepare yourself for actual operational playbook. 8:27 So this is kind of what I was talking about. 8:32 This comes from public research and papers. 8:37 These are still all well-known legacy agencies and companies that thrive on consulting analytics and market intelligence like Gartner's, McKinsey's, Forrester's.
0:00 Music 0:06 I want to show a use case and constraints where you can use OpenClaw and still have it securely and be sure that it doesn't leak anything sensitive on your site. 0:19 I'll explain why it's set up this way, what are the limitations. 0:23 So going like a little bit in advance. 0:26 Today, we won't give OpenClaw access to any sensitive information like to your email, to your contact data, to your parcels, credit cards, to your personal details or anything. 0:38 But we'll explore several use cases where it can be used securely and understand why and how to prepare it to further integration with other systems. 0:50 If you look around, there have been several experiments, and I've saw some thread on Twitter, now x.com, when someone set up their open claw, gave them access to 100 bucks and said, hey, make me a million. 1:04 And going forward... 1:07 It was like really spectacular to watch how this open floor signed up for different accounts, tried to trade stocks, tried to set up some sales funnels and benefit from like all the general knowledge out there. 1:22 But at the end of the day, $100 spent, nothing made, no profits. 1:27 So it doesn't work like out of the box. 1:31 How I like to explain it, AI amplifies the skills that you have, the talents that you have, and that mission that you bring here kind of to the human civilization. 1:42 So yes, anyone can do the app. 1:45 Anyone can repeat it. 1:47 But can they do it with the same level of attention to detail, the same level of passion and dedication? 1:54 Probably not. 1:56 So if you are on a mission, if you understand well what it is you're after and what it is that you want, 2:05 I think AI can help you to build it up the way that no one else will be able to repeat reliably with the same level of dedication. 2:16 Because at some point of time, the differentiation becomes not about what to build, but about the mission and the values that you deliver. 2:25 So just give your very high level example why I think so. 2:33 You kind of, if you look at the use cases of AI adoption in the industry, like overall, and this is like actually the perfect segue to our first slide in the presentation, 2:48 actually only 6% of the use cases 2:52 proved that they have any positive for a while at all. 2:55 So, like, extrapolate that to the market and the 3:00 product you mentioned built. 3:01 In 6% of the use cases, anyone trying to do the same can probably get some value of it and can make money. 3:09 And anyone can, but do they have that in themselves? 3:16 Do they have that understanding of how to build this together? 3:20 And why this fails to me looks pretty clear and obvious. 3:26 AI is compressed archive of all human knowledge that accumulated over the course of civilization. 3:35 So everything we ever know, everything in public domain, of course. 3:40 Everything that was ever put on the internet was compressed, categorized, systematized, and packed as like one huge archive. 3:49 And we can, now with school, we can ask this archival question, but eventually whatever answers we'll get are just permutations of different patterns matched to your question. 4:02 So you get out of this huge archive, basically some random permutation of previous patterns, some of which are useful. 4:13 And this is also kind of, I can remember my university professor or... 4:22 Was saying this thing about all mathematical models. 4:26 And his statement was, you know what, all mathematical models 4:30 models, extrapolation projections that we're going to build are on all of them. 4:36 None of them actually reflect the actual world. 4:39 But most of them that I will teach you are useful. 4:42 So in this way, whatever products, outcomes that you can build with AI, they are useful. 4:51 But they're always wrong. 4:52 And there will always be kind of this space dimension where you can fit in and find a way to extract positive value from deployment. 5:04 If you follow, again, your passion, your excitement, if anyone shares it, they can probably match you. 5:11 So it's like, for me, a game of people using this incredible technology to amplify and deliver to the world what they want and what they have. 5:23 Right now, everyone has tried chat GPTs, clouds, or any other solutions, co-pilots in their daily workflows. 5:34 And hardly anyone haven't heard about AI because all recent years it's been all around. 5:40 So they're like some hard mess. 5:42 This is just something that I quickly looked up online when preparing materials for this presentation. 5:51 Only 6% of the use cases. 5:54 Actually have and see some value. 5:57 And this comes to like, 6:00 different reasons to that. 6:02 And one goal that I have today is to equip you with the knowledge and understanding, first, why 6:10 it's not so common for everyone using AI to get the benefits and to have positive feedback and have the return of investment on their time and benefits. 6:22 And the other understands why it happens. 6:24 Today, what I want to do is first dive into some fundamentals and give you some very high-level basic understanding of the technology, of its limitations, of where it makes sense to use it and when it doesn't, 6:39 and also how to start using it for yourself responsibly. 6:43 Into some limited constraint and extent. 6:46 So in the second part of this day-to-day, we'll set up open core for like every one of you. 6:53 So again, like repeating myself, we're not going to give it access to anything like sensitive private, and I'll explain kind of why, but we'll explore use cases that will make it useful for you, like in your daily use cases in your daily life. 7:07 So you can't just ask AI, hey, make me some money. 7:10 Unfortunately, it doesn't work. 7:12 Maybe it will in a couple of years, but that will be a different story. 7:19 So in the nutshell, also another thing that is happening, like everyone is trying to use AI extensively. 7:26 And the trend in 2025 compared to the 7:30 previous year is also pretty obvious. 7:32 Only 70% of the projects across the whole industry were abandoned since 2024, but in 2025 and more. 7:42 It is pretty explainable because most of the experimentation becomes really low. 7:48 So it's easy to try things, it's easy to experiment, it's easy to build proof of concept. 7:54 Try your ideas and eventually fail to learn something new. 7:59 So from early failures gives an opportunity to validate your ideas, to try things, understand what works, what not, not to prepare yourself for building something that actually works and you can operationalize it. 8:14 That's also kind of another part today. 8:16 We'll kind of get these fundamentals to see how you not only can experiment, but how to prepare yourself for actual operational playbook. 8:27 So this is kind of what I was talking about. 8:32 This comes from public research and papers. 8:37 These are still all well-known legacy agencies and companies that thrive on consulting analytics and market intelligence like Gartner's, McKinsey's, Forrester's.