The AI Skills You Need Now to Stay Ahead With Joshua Wöhle

Heather Monahan [0:00] Hi, and welcome back. I'm so glad you're back with me this week. This week, you get to meet someone who I now call a friend. He's brilliant. We've got Joshua Bola on the show today. He's a six-time founder, AI adoption expert, and CEO of Mindstone, the AI adoption platform for non-technical. That's your girl, Heather. Professionals and home to one of the world's largest practical AI communities. Previously, Joshua co-founded Super Awesome, the world's largest kid safety technology platform. As chief product and technology officer, he scaled the business to more than 100 million in revenue. This guy knows what he's doing with technology protecting the digital experiences of more than 500 million children. each month across brands, including Fortnite, oh my gosh, that thing needs protection, Nintendo, Lego, and Warner Brothers. In 2020, Super Awesome was acquired by Epic Games in a nine-figure exit. Today, Joshua is best known for helping leaders and organizations like me move beyond AI hype and build practical AI capability at scale. At Mindstone, Joshua is closing the gap between AI's promise and what workforces actually experience. Named one of Time Magazine's World Top EdTech Companies in 2025. Mindstone has trained tens of thousands of professionals, including me, and including Delta Airlines, Home Depot, Hyatt, so many companies. The company's core belief is that most AI adoption fails. It has little to do with the technology and everything to do with the people, the habits, and the organizational behavior. Joshua has backed more than 40 startups as an angel investor and is a venture partner at Emerge Education. He regularly FTSE 100 companies and global innovation events and was a feature speaker at London Tech Week. 2025. I mean, I can go on and on about this man, but he's incredible, and he's here. Josh, thanks for being here with us today.

Joshua Wöhle [1:43] Thank you very much for having me.

Heather Monahan [1:44] Oh my gosh, quite the resume. Okay, take us back. First of all, everyone's freaking out because they're saying this guy's a tech genius. Take us back to your childhood, and did you know you were heading down this path? Did your parents know about this?

Joshua Wöhle [1:58] No, not at all. Um… I think it's… so I was… I always wanted to be a lawyer, actually, when I was, uh, smaller, and yeah, I don't know… don't know what went wrong there. Um, the…

Heather Monahan [2:10] Or what went right, very right, actually.

Joshua Wöhle [2:12] Oh, right. It was really when I, there was a class when I was 12. It was an extracurricular activity at school where they taught people how to use JavaScript, HTML and JavaScript to create websites. And it was like a one hour week, like on the side. And it just, since then, there's something about the feedback loop of creating something and instantly seeing it, the result of what you're creating on a screen and then being able to iterate. on it. That was… that has just forever kind of hooked me since. Uh, but I've always been more of a builder than I am… I'm not… I have a computer science degree, but I wouldn't say I'm… I'm not a computer scientist to do it. I'm a builder, and I use the tools that I need to build the things that I want to build. I've never been the best software engineer, but… Yeah, so did… grew up in the Netherlands, did about 13 years there, 10 years in Switzerland, and now 16 years in London, and um… Built a few companies along the way, and it looks like everything is accelerating now, so not stopping yet.

Heather Monahan [3:18] One of the great things about you is that you are a real-speak, normal person. When I had the opportunity to learn from you and get to know you, you make understanding AI normal and simple. And no one had done that for me. And I know for a lot of people listening, Josh.

Joshua Wöhle [3:18] Okay.

Heather Monahan [3:35] they're very intimidated by AI. I know you hear about it all the time. Where do you get people to start, to start embracing this? Because a lot of people are hiding and hoping it's going away someday.

Joshua Wöhle [3:47] Yeah, so… it's very unfortunate that we are on the back of a few kind of, I'd say, tech waves that have just not actually had the effect that anyone said was going to have. So, like, we had. blockchain, NFTs, we had, uh, the metaverse, and everyone always said, like, it's gonna change everything! And the people that said no at that point were clearly proven right. And so now that it's the fourth time in a row that. the techies are kind of saying, this is gonna change everything. Everyone's looking at it, it's like, yeah, we've heard this before, uh, seems complicated, I'm not really sure I want to go through that. Um, so that's one thing that's seriously hurting, uh, what is happening here. Um, now, in terms of the fear, the good thing that I can say is that actually the basics are fairly easy. I mean, we spent, what, three and a half days together, right? And… I think after day one. you already get a lot of the benefit of what the technology can do. It's not actually technology that's complicated. Once it's a bit uncomfortable. And actually, like any real learning is uncomfortable, even, I mean, for anyone that is interested in learning. You actually have a concept called desirable difficulties in learning sciences, this weird thing which is basically your neurons reconfigure themselves. And that's the part of where you are actually learning. But that is difficult in itself. If it is not difficult, you're actually not learning. And so that getting past that, the good thing is, is it's not that much of an effort. One day, and we actually, we do a four-week program, which is 10 hours in total. In 10 hours in total, people get back to 5.5 hours a week. So the good news is, even if you haven't started yet, um, you are still. way… if you start it tomorrow, you're still way ahead of the curve, and it's not… it's not going to take you months to get to the other side. It might take you months to get to maybe where I would be today, but, like, it snowballs from the first steps that you get through, and the first steps are easy.

Heather Monahan [5:58] One of the things that I — so guys, if I didn't explain this, I've been talking about it on LinkedIn. I did a three-day training with Mindstone, with Josh, in person, on site, in Santa Cruz. It was incredible. And not only did I — you're put in the room with unbelievable people from all over the world.

Joshua Wöhle [6:12] Those are awesome.

Heather Monahan [6:16] Very different backgrounds. Remember our guy, Dave. But we have people from literally — what kind of business does he have?

Joshua Wöhle [6:25] HVAC.

Heather Monahan [6:26] An HVAC business, which he has completely restructured, redone, and his company's blown up since he started working with Mindstone. He's reimagined that industry and the technology he's brought forward. And guys, he is a regular guy. I do not see him as a tech guy. But he's embraced the technology so much so he's taken the three-day training, I believe, two or three times.

Joshua Wöhle [6:47] Yeah, it was the third time, and he took his son along.

Heather Monahan [6:50] You brought a son this time. Right, and so… and they've completely reimagined the business such that when you have an HVAC problem, you hold your phone up, and… can you explain how… how he's doing it now?

Joshua Wöhle [7:01] Yeah, so he was building this system where his, um, staff, who were already professionals, um. When they don't quite know what the problem is, they use AI to help troubleshoot. You hold your phone up to it, figure out what the problem is, and get to the next step, as well as his customers ahead of time. It takes quite a bit of money for a professional. to come out and do the thing. If that professional can know ahead of time what is the likely problem, and they can have all the material ahead of time with them before they even get to your home, it means they come once, not twice, and that… it's a 50% saving for everyone. So, it, uh… yeah, it's… he's been doing more and more, and he's now playing around with, like, the idea of having these, uh, the Ray-Ban glasses, and having this real time for all of his employees and stuff, which is mental.

Heather Monahan [7:50] So he's just this regular guy. I would say he's less tech savvy than I am, and I am not tech savvy person. He's just such a regular guy, but he's completely reimagined his business, guys, as a result of going to the Mindstone weekend and in-person training. Here's what I'll say for me, and I know everybody learns differently, and I know, gosh, you can teach so much about this. This is what you do. When you have somebody hands-on standing with you versus in Substack, you can read about prompts and this. For me, that was the game changer because, Josh, I knew how to use AI. I had all these different assistant AIs here and Zoom AI recap this. But none of them merged together. I had Claude Cowork, as you know, and I was using that. I hadn't set up skills. I hadn't set up any automations. So I was still having to be everywhere to do everything. And AI was helping me, which was great, but I didn't understand.

Joshua Wöhle [8:35] Mmhm.

Heather Monahan [8:44] what you were just explaining to me, how you have 25 agents right now running on Rebel while we're here talking, doing… can you talk a little bit about, so people can understand the difference between basic AI use versus what you teach?

Joshua Wöhle [8:58] Yeah, okay, so, well, there's what… and it changes a lot, so even what we teach now will be different in two months from now, right? The, um… so, the basics of AI, we kind of think about it in different levels, so… Level one is you're you're using it. like a search engine. Level 1, you're using it just like an answer engine. You ask a question here and there. If you get familiar with it, people start talking about how they use it a lot, it's replaced their search engine. That's, uh… and often people top out there, because it feels like, well, I'm using it all the time, surely this is what it is, and… And that's also where the underwhelm really comes from, because, you know what, Google was a decent search engine. If all you're getting out of, like, a ChatGPT is that it's a slightly better search, then you're getting some value, but it's not revolutionizing. The kind of third level becomes using AI as a thought partner. And that's, I think, where we started most of the weekend with, which is, okay, well, what if you stopped asking it questions and instead you ask AI to ask you questions about the problems you have so that it helps you think, which also removes a lot of the negative aspects. Some people are saying, like, it's. removing your thinking, it's making us dumb. Well, actually, if you get the AI to ask you questions, it makes you think more. And if you do AI well, I can tell you from experience, like, I have days where I get to 11am and my brain is fried. Why? Because I've been using it way too much, and the only thing I'm doing is thinking. Um, so once you get to past Thought Partner, you then get to what you would call the Agentic AI layer. There, we talk about AI as a teammate. So this is now, not only is it asking you questions, it's second-guessing, it's giving you extra details, but it's actually also able to act. on what you're asking it to do. So now, it can't just help you build a proposal, it builds the proposal for you, it sends it to the right person at the right time, after having done all the research on them, um, it… communicates with other members of your team, trying to gather the information they need, maybe in order to do their job, or to get a daily update back together so that everyone in the company understands what everybody else is doing in the company. And it's able to hook up to all your systems, whether that's your CRM. your Gong, your Zoom, your email, your WhatsApps, everything that you kind of use in order to be that teammate. And then there's a last level, where I would say, or I'd say last level, no, that it's kind of the currently. edge of where you can play, which is AI as an operating system. And this is where you no longer just have agents, you have teams of agents, and those teams are all orchestrating with each other, maybe with an orchestrator even on top, and that is what I'm… what is happening right now. So, like, on the second screen at the moment, I've got 25 different agents. that are… well, actually, 24 agents and an orchestrator. The orchestrator basically gives feedback to the other 24, uh, in order to ship product for us 24-7. Uh, and so it's literally going through bug reports that users all over the world are real-time emitting. based on their own usage. It just picks them up, squashes them, redeploys a new version, and everything is done. And I won't even have known about it, let alone, whatever, maybe 20 of them are big features that we just wanted to add to the product, and 4 of them are bug squashes at the moment, kind of coming through. And the same is happening now at a non-technology level. So, we were talking just before, but the… I've always… not just felt, at this point, I would say the pattern has been so often that software engineering kind of lifts the future a little bit. It's just the… the window of the future is getting smaller and… like, the distance between software engineering and everything else is getting smaller. But before the end of this year, you'll have this same concept. It's called… in software engineering, people talk about the software factory. And there's kind of a same… the same thing is going to happen with every other profession, or most other professions, anyway. So you'll have a legal factory, a customer success factory, a sales factory, and these will be… Teams of agents that are just proactively doing the work, surfacing it to humans when humans need to be involved, and otherwise just doing it completely autonomously.

Heather Monahan [13:27] Okay, so some people just listen to you, and I know you, I trust you, and I've learned from you. Other people haven't, so they're getting defensive right now, I can just tell. That was intimidating. To that end.

Joshua Wöhle [13:32] Yep. Yep.

Heather Monahan [13:40] can you explain how anyone can make it such that it doesn't have to be autonomous, that I can set it up so it checks in with me, and how you have all the permissions that you determine?

Joshua Wöhle [13:46] Yep. 100%. No, this is… this is absolutely a journey, and also, like, I didn't get to the point where the agents would run autonomously until I'd spent months, basically. telling it, or going through the permission system where I would have to, uh, tell it yes all the time. So the, uh, your assistants tend to not do anything without asking you very specifically. Actually, the default, even in co-work, uh, the default until a week and a half ago. was that it would ask you for everything. I think since a week and a half ago, they flipped the default to, um, uh, always accept, or, or, uh, auto-accept, sorry. The, which means that the mo- the, um, the AI itself decides when an action is important enough to surface to you. in order to ask for permission. So, for example, if it's about reading your email, it'll say, that's fine. But before sending an email, it's like, oh, if I do this, it's a non-reversible thing. Someone is going to get this message. I might ask the user first. And at a genetic level, that's the, I think, the right way to do it. Umm. you might think that just getting… switching on manual approval would be the better way, where it asks you for everything. The reality is, if you get it to ask you for everything, you're going to ask… it gets so many requests, that the only thing you do is say, yes, yes, yes, yes, yes, yes, and actually, you will click yes on the one thing you should have said no to. But because you just did 50 yeses, you're getting in auto mode because going through, so it's much better to just get it to only flag the ones that are actually somewhat sensitive, so that you still really take the time to think about it. But all of this is fully configurable, and when I say that. By default, if you use something like a Rebel or you use something like a Cowork, all of this out of the box, anything that's sensitive at this point will get flagged to the user. So it's very rare that any of these products would do something you didn't ask it to do. Which is very different from an open claw that you might have heard of or an Hermes, which it's more on the edge, but all of the disaster scenarios often come from those environments where specifically software engineers being a little bit on the edge decided to just go all in. and suddenly it deleted a thing it was not supposed to delete and stuff. That doesn't really happen.

Heather Monahan [16:11] So for people who don't understand this, and I'm super interested to hear what you think, two things. How do we know what to use? Rebel, Claude Cowork, ChatGPT, Perplexity, Gemini, there's so many different things to use. So how do we know what to set up and what to prioritize? And then once we know that, how do we know what level, because I keep getting notices now, Fable 5.1 is on. Those two things confuse a lot of people.

Joshua Wöhle [16:16] Okay. Yeah, so for most people, you can tune out almost everything. I would even argue start with ChatGPT or Claude. I mean, those are the two best starting products. Get familiar with getting it to ask you questions. That tends to be a real unlock already for people. Use voice, so talk to it, don't write to it. And when I say use voice, use the actual voice mode, not just the kind of the voice to text feature, but actually go into voice so you have a real conversation with it. Um, start there. Go deeper there, and give it things that you think it is impossible to get right. And you'll be surprised how much it gets… it actually gets right.

Heather Monahan [17:22] Oh, my gosh. So you were teaching us a couple of things that I want to highlight. Text is the lifeblood of AI. And I didn't understand that until you taught me that in our time together that weekend. And that changed everything for me. Every client I deal with now, I record everything. You actually wore this pin the entire weekend, even if we were having dinner at night, anything that was going on. And you would record all the conversations so that we didn't miss anything, so that we had summary notes, so that we had action items. So when we came up with great ideas.

Joshua Wöhle [17:26] Mm-hmm. No. Yes.

Heather Monahan [17:52] After dinner, we actually had a game plan the next day. It was unbelievable. And so I've taken that really important piece that you taught me. Even if I'm doing a potential, I'm going to potentially work with someone. I tell them I'm recording this conversation not to share with anyone other than you if you'd like it.

Joshua Wöhle [18:00] Yeah.

Heather Monahan [18:09] I'm using it to make sure I don't miss anything. I can't tell you, Josh, how many times that I will bring that information back into Claude Cowork after I create a file on the person, right? I store everything in Google Drive, just like he taught us how to do, and everything's got a folder to stay organized. I'm done with this meeting. This is a $60 billion client. He's the COO. I have this great call, obviously huge client for me, and um… I need to get him, by the way, to the 3-day weekend, so we need to discuss that when we hang up, because he's got to come. We've got to do this. He's gonna love it.

Joshua Wöhle [18:38] That'd be awesome.

Heather Monahan [18:39] It's going to be incredible. And so here's the thing. It said to me, Heather, you missed something very important at moment 2233. And I thought, I'm a great salesperson. What are you talking about? And so I go and actually, and of course I have the Zoom recording and I go to watch what I missed. And he was right. I missed — he populated something. I was so — and this is embarrassing to say probably, but I'm just going to tell the truth. I was so jumping to the point that he had surfaced something, and I was thinking instead of listening in a moment. And I missed in one moment a huge thing.

Joshua Wöhle [19:02] Hmm, yeah. Yum.

Heather Monahan [19:13] Huge important piece. And I would've forgotten it because I wasn't truly listening. I was speaking over him and Claude Coburg caught it. I was able to send an email after and apologize for speaking over that and not highlighting it and circle back on our next call. And that's when I actually closed the business with him.

Joshua Wöhle [19:21] Yep. That is awesome, indeed. And you're totally right. I mean, transcripts are such a dramatic boost on AI. It's the… Um, especially when you have a system, or you could do it manually, where you're well organized. So the… if you can point the AI to the point where you say, okay, hey, here are all the transcripts with this customer. Now help me figure out how we can put the best possible proposal together." It is so much better than at least I have ever been. Why? Because it's impossible for me to keep more than even one good conversation fully in my head so that I don't miss details. The AI doesn't miss the details. And it does this across five, six, seven conversations. So when you have to do really kind of big reasoning, whether that's for a big strategic proposal or it's internally because you have a big project that you're working on with five or six different people. taken 6 meetings to get together, like, trying to condense that into something the whole group can agree on is so dramatically easier when you have the transcripts and working with AI than if you tried to do that yourself.

Heather Monahan [20:40] One of the other cool things that you taught us, and then I'll stop being this dead horse around the text portion, we can move on from it. But so when you taught us that we have the text as a like button, once we have that, everyone learns differently. And so sometimes there are different settings for different ways to consume information. And a cool example you took us through is you did a recap of our prior day in like a podcast format because you fed that same chat thread or the same text thread that was the actual transcript from our day the day before into, I don't remember.

Joshua Wöhle [21:12] It was NotebookLM.

Heather Monahan [21:12] Well. NotebookLM. Thank you. And so you fed it into NotebookLM and then you picked podcast and you made a podcast, like a 10 minute podcast out of our previous day. And it was so cool. Everyone was so highly engaged. And so you start thinking, wait a minute, if I could do that, how could I?

Joshua Wöhle [21:15] Yep. Yep.

Heather Monahan [21:29] How could I make learning entertaining or recapping meetings entertaining for my team, for my employees, for my clients? Everyone learns differently, and now we have the ability to literally take that same information and make it anything we want.

Joshua Wöhle [21:44] Yeah, the multimodality piece is completely under… underused, I think. And I include… I must say, I include myself in this. I'm probably already… I'm using it more than most. We're about to go and do something really interesting. So the… I won't divulge too much, but we're working with an author. And we're going to be using a lot of this enhancement technology to enhance a book, which is a really interesting way. So now you'll have a book, and the book… has, like, an accompanying AI, basically, that holds you to account as you read through the book. Um, so, like, think about what this is gonna do to, like, the self-help type category, which is anything where you are trying to improve. yourself, any of the books that teach you something. Now you can. it'll know where you are, how you go through the book, and how you should apply what you're doing as you go through your day. It'll literally just be able to tell you, by the way, the thing you read yesterday, today you've got this thing coming up, and you should really be applying that.

Heather Monahan [22:51] That's incredible. And please share more as you develop this because as you know, you are sitting here with an author who's working on our third book right now. So I definitely, I need more information on that one. Thank you. Okay.

Joshua Wöhle [22:59] Yes. Oh, we should work on that.

Heather Monahan [23:04] We need to work on that. So, for everyone that's listening, and I know you hear this a lot, um, AI's taking our jobs. I know that you say someone who understands AI and uses AI more than you do could take your job. What can people do now to start setting themselves up for success for tomorrow and for 6 months from now?

Joshua Wöhle [23:25] I mean, really the only thing you… can do is… is to lean in. Um, is… is to just start now, and… and the good thing, as I said before, is that you're actually not behind. So, um… Just start… Put your toe in the water. use an AI system. Use it personally, get it to ask you questions. Throw problems at it that you don't… think it could help with, and then try and second-guess yourself a little bit. It's like, try it, and it's like, what if I tried? Um, and then maybe get some, if you can, more and more companies are now getting support. If you're in one of the companies that is properly training, make sure you take advantage of the training that is being given. to you there, because what I can say is right now, there are still quite a few, not by far actually the majority of companies, it's still okay not to be using this technology. In three years from now, it's not going to be. It's going to be very similar to trying to have a job and saying, sorry, I don't use email. So you can try, but very few employers will be okay with that. And I do think the longer, not just think, I know that the longer you wait, the gap is increasing, not decreasing. The difference between the people that really knew how to use this technology three years ago and the people that didn't. was much smaller then than it is today. And so, the leverage you get out of starting early is really the thing that changes the game.

Heather Monahan [24:56] So for somebody that wants to go through what I did, the in-person three-day, that they're saying, you know what, I need that hands-on, someone standing over me and actually helping me because there were times I couldn't find something and, you know, Jeremy had to come over and reset my computer and restart things that I just didn't know to do.

Joshua Wöhle [25:04] Mm-hmm. Yep.

Heather Monahan [25:13] Um, and it was… I can't even say how helpful it was for me. It allowed me to break through. For someone who wants to learn like that, who are the people that should attend your 3-day that you have coming up in October?

Joshua Wöhle [25:26] So I'd say there are two types of people. There are… the CEOs, or the C-suite of some of the bigger organizations who are thinking, okay, I need to understand how much of this is true, how much of this is not, how much of this is hype, how much of this is actually possible, so that I can then go back. And know what to even do. demand-slash-ask from my team to begin with, right? To know that I'm not asking an impossible thing, because I've lived the actual, uh, possible. That's one category. The other are smaller… we have quite a few… owners of, say, mid-sized companies who come, and they bring a few of their staff along, and then it's like, okay, well, we're going to go do this, and we're going to do this as a team, and we're going to live that, and then instantly go back, and then each one of those execs ends up rolling it out across their own team, and basically leading by example in that way. So I'd say those are probably the two good ones. And then you have a few freelancers here and there, like basically people that work on their own for themselves. And they're just, okay, I've got a team of two, three, four, maybe, and I need to operate as if I were 50 or 100. That… we've got a few of those that come through as well. Basically, they have a lot of agency, and they can decide what tool to use, how to use it. They don't have any of the constraints of the big business, where security might ask them XYZ. They are able to go in and say, okay, I need to go do this, I'm going to move this this week. Just be hands-on and live what's possible.

Heather Monahan [27:04] One of the things I really liked is that you focus on having tangible takeaways at the end, that you wanted systems, products created. I created a platform. I can't… I've created websites since I've left you because I… and again, I am not a tech person, but you taught me how simple it could be, and it just keeps getting easier and easier now. Do you have any examples of some of the business owners that are mid-size or large-size companies that you've worked with and how they've had success after they've left?

Joshua Wöhle [27:32] Yeah, so we have one person, runs an accountancy firm, I think they're 15 people, and they came to one of our weekends almost a year ago, and then sent another exec to the one after that, and then a third. the one after that. So, like, that was, uh… and they were seeing dramatic results for themselves individually. Let me just think about on the smaller side. Oh, we had a surgeon at one of our… one of our weekends. This was in Florida, and they went through how they… how they evaluated their own practice and their own research. And they said it had completely changed their ability to. think through particular problems in a way that they had never been able to do before, and I have some emails come back on, like, what dramatic effect it had, like, life-changing, and I say, when I say life-changing, life-saving risks with patients that they were able to go through afterwards, like, so… You have on the small side quite a bit there. On the big side, you met a few that were on the weekend that we had there as well. But we've had CEOs of companies, 100,000 people and up that are just, okay, you know what everyone's telling me? that this is a thing. I don't quite get it yet. I don't know how much of this is hype and how much of this is real, so I'm just going to have to get my hands dirty and spend two and a half, three and a half days going really deep in this, and then go back.

Heather Monahan [29:09] So I think I might have shared this with you. One of my dear friends is the head of a major bank and he's the number two guy in charge and he has never used AI. And so I stayed the weekend with he and his wife and their kids a month ago, right before I came to your three day and I was already using AI and I was floored that he is running such a massive organization. All of his employees are using AI and he's not.

Joshua Wöhle [29:11] Mm-hmm. I remember.

Heather Monahan [29:35] And when I said then, I don't understand, what are they doing with AI? He said, I have no idea. To that person, and I couldn't get him to understand, Josh, gosh, how are you leading if you're completely in the dark? I'm not saying you have to be the best and the most proficient, but I don't understand how you can lead a team if you really don't even know what they're doing.

Joshua Wöhle [29:48] No. Yes, and that is… fundamentally misunderstood at the moment. And it's misunderstood for a good reason, because up until recently, technology was the role of the chief tech officer, chief technology officer. And, um… the… I was CPTO at Super Awesome. I was very much… I would always… I used to take pride that I would… I basically would be able to tell you what was possible, what was impossible. Like, that was something I was really good at. Um, no longer can do that, because this is such a dramatically different technology. And actually… It's not a tech problem. This is a very easy to use technology. Um, if you're not getting the ROI, it's not that the tech isn't capable of it, it's not that users don't understand how to put text in a text box, it's that they're still working the way they used to work, when really what they should be looking at. is thinking, rethinking how they could do a job, their job, now that intelligence is so easy to access, now that they have a co-worker that sits next to them 24-7. Most of us, like, most people in the world have. never managed a person, have never had a chief of staff. I mean, there's very few people who would have had a good chief of staff in their life. And it's actually not that easy. The first time you have a chief of staff. figuring out, wait, what do I even give them? How do I give them the context for them to be able to do a good job? It's a really frustrating job sometimes for a chief of staff, because they're trying to help you, and, like, they don't have your email, they don't have your messages, they don't have all your relationships, and you're somehow supposed to help. The chief of staff is somehow supposed to help you do your job, but they don't have the context where even the best person in the world would be able to. Weirdly enough, the same thing is true for AI. And so if you don't give the AI enough context, you don't give the AI enough tools to be able to do a job. It can't do the job. And so the… when people… think this is, again, just the next technology wave that is just going to be the responsibility of the CTO, and the CTO puts this in their bucket of technology, it is going to be applied to all the problems where technology has been applied in the past. This technology is fundamentally different. For the first time, I don't want to go too technical, but there's two types of technology. There's technology that is predictable. you put something in, you always get the same thing out. And this technology that's non-predictable, it's called non-deterministic. Basically, you can put 100 times the same question into ChatGPT, and you get a slightly different answer out of it, which is actually very similar to people. If you ask me 100 times the same question, at least, well, unless you ask me 100 times in a row, it's unlikely you're going to be getting a single time the same set of words that come out. It's going to be directionally the same. but the actual words that I'll use will change. And so, because this technology is so different, all of the problems where you try and apply it are actually fundamentally different to anywhere you will have applied it before. So, almost by definition, when this becomes the problem of the CTO. You end up never exploring the set of problems that this could have solved. And as a CEO, until you live yourself how this is useful to you. It's really hard to communicate how the rest of the organization should think about it. The second piece I would mention. is that the mere fact you're not using it is exactly the sign that you don't understand it. Why? Because this is… decision leverage. This basically gives you intelligence on tap. And if you as a CEO say you don't know how to leverage the most intelligent people in the world to help you make better decisions, you have a bit of a problem. Um, and… You as a CEO of a big organization, your decision leverage is the highest it is in the entire company. Your decisions have the biggest direct effect on how the company is going to be moving. So if anyone in the company should be using. AI to amplify their decision-making, it is the CEO. And if you're not using it, it just fundamentally shows you don't know what the technology is about.

Heather Monahan [34:27] Oh, that's so true. Oh my gosh. For everyone listening, you have got to attend this three-day event. It's coming up in October. It is going to change your life. Josh, tell us a little bit about what does the next six months, one year, two years, five years look like with this technology?

Joshua Wöhle [34:29] Yeah. I don't think anyone can think 5 years. It's moving so fast. It's actually scary to extrapolate to 5 years, so I can maybe say, like, 6 months. I can talk… I can definitely talk 6 months and 2 years in terms of how the world is going to change, because the reality is where the technology is today. most people will only find it out in 2 years or 3 years. Um, so the… From a technology perspective, um… we are likely going to be hitting this year, and I'm not the only one saying it, OpenAI is now saying it since this week as well, we're likely going to hit the point where. that AI is going to make better decisions than the entirety of, uh, humanity combined by the end of the year. In terms of capability, which means it'll… uh, well, it already has discovered new science, but it'll start to discover new science that is going to fundamentally change the way that we… Understand the universe that we are able to build, like, we'll be… there'll be an acceleration in productivity across the board in a way that we have never seen before. Like, we'll be able to suddenly, over the next few years, we'll be able to. Cut the time it takes to grow food, cut the space you need to grow twice the food, you'll be able to build houses faster, you'll be able to find material that resists better so that houses don't fall down as easily. All of those things are going to be a direct result of science moving on and AI is getting really good at help at the moment. at accelerating how fast humans can do that, and by the end of the year, we'll start to self-discover these new, uh, um, these new, uh, areas in science. Um, and… At a… for more kind of the everyday person. you'll hear a lot more over the next 6 months to a year about this Agentic working piece, so having the AI actually perform actions for you. That is the big thing that everyone will hear about. Everyone… Many companies will start to roll it out, so that it'll be able to help you do your job more, and… Uh, that… if you accompany it with some decent training, will have a direct effect. We're seeing… We're seeing anywhere from, kind of, half a day to two and a half days back per person in terms of their actual workability, which means that half of your week. you can now dedicate to other things. Most of the time, companies will find very good priorities as to where you should be spending the time, so I'm not saying this will automatically end up with everyone working less, that's not quite… but in terms of how your work is going to change, it's going to be. pretty dramatic. Um, well, it'll be interesting to see… There's a big question mark in the financial markets around AI, which is, will this productivity… Um, so the productivity increase is absolutely real, and it'll be there, but we have this interesting, weird thing where you have Anthropic and OpenAI, which are the two really big companies, they've raised a ton of money, and there's a question about if they will be able to make a lot of that money back, because you also have a lot of. open-source free models that are now coming in the market. And actually, most people that will be listening here. You don't need to use Anthropic or OpenAI models. Like, actually, you don't need to pay for it. Um, because unless you're an expert user, and you need that really high level of intelligence, most people can't take advantage of that right now. Maybe if you start today, in 6 months to a year, when you start to get to the more sophisticated pieces. you can start to take advantage of it. Today, probably not, which means you might not have to pay for it. And so it'll be very interesting to see, because both of those companies are going to go public, I think, before the end of the year. to see if those companies will be able to make the money back. Depending on that, the markets are going to take a positive or a negative signal, and we can definitely be looking at a weird pattern where the productivity increase is very real, but the markets still crash because OpenAI and Anthropic are deemed to not be able to make the money back, which is now… I think Anthropic is supposed to go public for, like, $2 trillion. Which would be the biggest IPO in history. So lots of activity over the next six months alone. And then if you kind of think about in a year from now, two years from now, it will be… completely normal to consider AI as part of the org chart in the organization. So, I had a dinner last night, uh, it was with, what, 10 C-suite, different companies that we hosted. One of them had already, today. AI employees are part of their org chart. When they communicate their org charts, there's just, okay, we have… and they give them names, and so Johnny Customer Success is part of the Customer Success team. But… and they do that because. what I was talking about before, which is the… that means that that AI is onboarded. It means there might be an off-boarding process if they decide to decommission the AI in different ways. It means the AI needs to be in the loop when there are discussions, when there are team meetings, things like that, because, again, the AI can't do its job if it doesn't have the context that the rest of the team has. So, all of that will be absolutely normal. Um… And I think one of the things to… maybe one of the downsides, which is also interesting to look out for, is that we are all going to have to deal with the idea that if you no longer do any of the grunt work. Which is something that I had to deal with myself, and we actually did a fireside chat on this at the weekend in Santa Cruz. You have this thing called, kind of, brain fry. Um, which is just something I definitely haven't had to deal with, and many people that have gone through Agentic haven't had to deal with, which is, um, if the only thing you do during your day is making decisions, your brain is not tuned, or at least my brain was not capable of staying normal if the only thing I do is make decisions. I hadn't realized how much of my day was make a decision, do a bunch of work, make another decision, do a bunch of work. If the only thing I do is make decisions, by 11am, my brain was just gone. I couldn't understand, but I couldn't think straight anymore. Um, and so you're going to have, or we're all going to have to find slightly different ways of working to allow for the fact that you just can't be making decisions the entire day. You're going to have to find other activities that regenerate your ability. to make decisions, um, and make sure that the agents are doing… are doing their job. So, I'd say that's kind of as far as I… as I can go, and it becomes really, um… kind of La La Land-y when you go two years and further. It's really hard to predict. I don't think there's anyone that can genuinely predict, even the people in the AI labs, they try at the moment, because it's good to give the world a bit of an idea of what might. But the speed of progress is so fast that no one can really say.

Heather Monahan [42:08] Can you give a little bit more insight into what actually happens in the three-day? And you even said to me that this new three-day in October is going to be different and cover different things than we even covered just a month ago.

Joshua Wöhle [42:22] Yeah. So I mean, the general —. Kind of the overview will be similar. Uh, which is, you start with… You start with. a good view of what is possible. A live demo of, okay, let us show you how deep the rabbit hole goes, right? What is possible today? Full-on live demo, no slides, just… Live in the room. We then go into setting up your own agent and feeding it, starting onboarding your own agent, feeding it with the context about you, your life, and then start to use that agent to work on a problem that is directly. relevant to you at that point in time. So that it's… the idea of the weekend is not only that you understand what good looks like and what's possible, but that you actually will have done more work in those two and a half days coming out of it than you would have been able to do had you not come to the weekend. So the idea is you don't waste any time, you actually gain time. and you're better for it. So you work on a real priority that is part of the business. We then go into what we would call shared memory, um, so we go into. What happens when agents are deployed across the whole organization and they start to collaborate, and how you can take learnings that one person in the business makes and suddenly helps another person in business do a better job instantly as they. are able to go for it. We simulate this by all people in the weekend, like, they're all basically building something together. Um, then it's building for non-builders, so this is about the software engineering piece. Um, so this is one of the other big pieces we didn't really touch on, but it's this, um. Um, so we fundamentally believe that the ability to build software for your job is soon going to be just as normal as the ability for you to manage a simple Excel sheet. Um, and it can be an extreme unlock when you can build simple apps for yourself that help you do work faster, just like you would have spun up a simple Excel sheet to sometimes keep track of things and make sure you have your own kind of little dashboard, stuff like that. Then we go into how to lead. So basically, how do you take what you have just gone through and you have just built for yourself, the unlocks that you went through, what worked, what didn't work? How do you take that back to an organization? What are the. blockers you're almost certainly going to be running into? And what can you do about making them slightly less big? Um, and then the last thing is we go through, um… a roundup, which is have your 30-, 60-, 90-day plan. Every day at the end of the day, actually, we do a little bit of a recap, and we take everything that you'll have done throughout the weekend, so including all your transcripts, all the problems you worked through, the actual curriculum of the few days, and then you put that together into. something you can take away that holds you accountable. We now also do automatic… the AI will actually hold you accountable, it will place the calendar invites in your calendar, it will call you at the right times after the weekend to make sure that you keep accountable to the problems that… or to the commitments that you had made. So you have kind of this digital accountability buddy that, um, that helps you actually implement what you decided was the right thing to do on the weekend directly, so… That's kinda where we're at now.

Heather Monahan [45:49] It's mind-blowing, and suddenly I feel like I need to go to Scotland. So for everyone who's listening, it's a Friday, Saturday, Sunday. So this is not taking them out of the work week.

Joshua Wöhle [45:57] Okay. Well, it's a Thursday afternoon to Sunday morning, yeah.

Heather Monahan [46:04] Thursday afternoon to Sunday morning, October, and it's in Scotland. Where in Scotland?

Joshua Wöhle [46:09] It's at the Orton Estate, which is, um, it's in the middle, so you fly either on Aberdeen or on Inverness. It's in the middle of Speyside, which is, uh, the world's biggest whiskey country, for what it's worth. Uh, they also do salmon fishing there, for anyone that is, uh. Interesting that I'm actually going a day early and I've never done this in my life, but I'll learn how to fly fish salmon, so we'll see. It's a beautiful estate. It's going to be about 35 of us, so it's very, very limited because obviously you want to make sure that we can really give attention to everyone that is there. It's right next to the River Spee. We have a beautiful location. The whole idea is a little bit of the clash of being on the absolute frontier of technology and then being in the most beautiful natural surrounding. Uh, so that if people need to see a little bit of green because it's becoming a bit too much, they can just step outside and, uh… Take a breath of fresh air and then continue.

Heather Monahan [47:11] So if people have questions, they want to know more, and they want to be in Scotland and understand AI, finally, and have someone teach them hands-on, how do they sign up? Where do they go?

Joshua Wöhle [47:22] So go to Mindstone.com, and there's a menu directly there on AI Breakthrough Weekend, and you can just sign up straight on the website. There's a brochure that will get sent to you as well, and then, yeah, let us know. There's still a few slots available, so I can also just… Email us directly, so hello at Mindstone.com and we can take it from there if that's easier.

Heather Monahan [47:42] All right, that's Mindstone.com. It will be in the show notes. You've got to make this trip, guys. I understand Scotland might not have been what you thought you were doing in October, but this is what you've got to do in October if you want to understand AI and you can't afford to miss it. Josh, how do people get ahold of you?

Joshua Wöhle [48:00] Uh, so I'm mostly on LinkedIn, that's probably good, so follow me on LinkedIn, I'm also on Twitter, well, X, um, and uh, yeah, and… Come and say hi.

Heather Monahan [48:09] Well, thank you so much for the work that you're doing to make AI something so complex, so simple. It was game changing for me, and I'm grateful for the work that you're doing and grateful for you being here today.

Joshua Wöhle [48:20] Well, thank you for having me.

Heather Monahan [48:22] All right, guys, keep creating your confidence. You know I will be. I'll see you next week.

This transcript was generated automatically and may contain errors.