Decades Change

>> Anouk: Welcome to fusion talk with anouk and steve. We're recording.

>> Steve: We are

>> Anouk: recording. It seems like ages.

>> Steve: It's been a while.

>> Anouk: It does. But that's okay then. Yes, we've been busy. Summer holidays.

>> Steve: Summer holidays. Life was getting in the way of all of it.

>> Anouk: So dead right. It's not bad a thing though, is it?

>> Steve: No, not at all. But you have something to celebrate?

>> Anouk: Uh, yes, I guess I do. Uh, I have a new position that I will be starting. Starting. Actually, uh, tomorrow. Yes, I might be getting nervous now.

>> Steve: That will be a first.

>> Anouk: Yeah, yeah. I start with a company in Belgium called Kronos. Um, I'm not going to go into too much details for the time being, but it's uh, it's uh, with the group rather than a particular company in Kronos. And so, yes, that's me tomorrow.

>> Steve: Exciting times ahead. But it will be fun.

>> Anouk: Oh, it'll be fun. It'll be a challenge. Uh, and definitely exciting. There's no doubt about that.

>> Steve: You need that challenge.

>> Anouk: And you've got some good news.

>> Steve: Yes, I'm not going in any details yet until I have the contract.

>> Anouk: But it's a change for you regardless.

>> Steve: It is.

>> Anouk: So it goes from more doing to leading.

>> Steve: It will.

>> Anouk: That will be fun.

>> Steve: Yes. And it will be a challenge and everything, but yeah, mhm.

>> Anouk: Yeah, I've been there, done that. When you have teams all over the world kind of thing, it's um, yeah, it is a challenge. But we'll get through it and everybody will phone and help you. Anybody that's listening on the podcast, if you get stuck, just give them a call.

>> Steve: Be careful with what you are saying now.

>> Anouk: You might get what you ask for. That is true. That is true. So the world is changing so quickly then. So what's changed over the summer?

>> Steve: What hasn't changed over the summer? I'm um, guessing if you look business wise or work wise, AI is going to be even more bigger and bigger and bigger than what it already is.

>> Anouk: Yeah, but we had this conversation earlier about how long it's actually been around and about how much has actually changed. Actually really, really changed. So I was doing this. I have a presentation to do on Wednesday. Next Wednesday? Yeah, I'm doing smart bites here in Belgium. I think it's Wednesday. Whatever, one day next week. Sorry. Yeah, but I have got the data.

>> Steve: You're sure it's next week?

>> Anouk: No, but I think so, yes. It's got to be because the week after we're in Zagreb, so I know it's not then. Um, and, but I was, I was. I'm basically doing, uh, a bunch of stuff around the gifts of AI that we've had. And I was surprised the date. So you. You got it right. First time, five years. Um, the AI has been around, or in terms of, we should say CHAT GPT popular. Do you know how long it was before ChatGPT appeared and Copilot appeared,

>> Steve: and what was before AI and all of

>> Anouk: that when ChatGPT was created and then Copilot was created. So Copilot obviously is AI from Microsoft. But you know how much time there was between ChatGPT and Microsoft? Five months.

>> Steve: I was, uh. You were too fast answering the question.

>> Anouk: You're too slow. You were too slow.

>> Steve: I was going to say six months.

>> Anouk: Oh, you'd been wrong anyway. But yeah, yeah, it's quite praising. So obviously those plans were obviously in place beforehand.

>> Steve: Yeah. But how long are they building on AI before it was released?

>> Anouk: Um. Oh, I mean, I think, uh, I don't know. That's a really. Another good question. How long was ChatGPT in existence before

>> Steve: it released the product in 2018 or 2008? I'm not sure about any. Uh, the exact date they were working on it already.

>> Anouk: 14 years. And one would be four years, according to Ask ChatGPT. Ah, because you're interested now.

>> Steve: Yes.

>> Anouk: Okay, we can do that. Well, we were, we were thinking about the change aspects and the human sides of AI, because everybody's been talking about it and we've heard so many stories, um, around this where, you know, the human, uh, edge and all that kind of stuff. What does the number.

>> Steve: It's OpenAI is founded in 2015.

>> Anouk: Okay, so 10 years.

>> Steve: No, uh, sorry, 11 years.

>> Anouk: Yeah. Seven years before they release the product.

>> Steve: Yeah. So they are building for 11 years on it.

>> Anouk: Seven years before they released the product? Yes, they released the product. 22.

>> Steve: Yeah. And here sets. If you mean OpenAI especially, the company was founded in 2015, so they have been building modern AI SIM SIMs for about 11 years.

>> Anouk: Okay, so either you're being completely dumb or you're not listening to me. The question was not how old they are. The question was how long were they building AI before they released it? So I guess you. Misunderstanding I do understand, but yes, seven years from creation to product. It's not bad, is it? Anyway, we wanted to think about the change because the stories have been hearing like, you know, um, uh, Amazon got rid of 30,000 people and then they brought them all back because AI was too expensive. We know that Microsoft changed their teams around, but the Numbers that they let go, they brought back a lot of people because AI doesn't appear to be delivering the goods. So we don't have an answer for everybody. But we thought, what level of change have we seen when other significant, ah, areas happen, especially in business and work. So we thought, I wonder what it was like in the 50s and the 60s, 70s, 80s and 90s and uh, 20s.

>> Steve: Yes.

>> Anouk: So we asked some AI questions and started looking. We come up with a bit of a model. So we're going to look at 50s, 70s, 90s now and maybe dig into the future. Not that we know what the future is going to hold.

>> Steve: We don't know what the future is going to be. But it's just fun to speculate about it and see if we are right or wrong.

>> Anouk: No, I agree. And I've done a few presentations on this before where you look at. There's actually, uh, a development path where you can see work changing and the way it works. So we're going to do a subset of that. We're not going to dig into it. And there are also people that invented new work. So for example, we're going to talk about the 50s, so we'll get into it in a minute or two. But that concept of, you know, owner, uh, manager, employee, producer kind of scenario, it was never like that. Before the war. Before the war, it was all just a small team of people doing it. The owner with whoever was helping him and they would get on with it. There'd be no structure. But after the war, there was some concepts around business and structure and so those came into play. So we're going to kind of jump in at that level.

>> Steve: Yeah. Do you know what's funny about this? We are Talking about the 50s, the 70s and the 90s now and the future. I can only understand now and the future because I didn't work at those other periods of time.

>> Anouk: Thank you for the reminder. Right, I am really gonna give you some shit for the rest of this podcast just. Just because.

>> Steve: Because you get.

>> Anouk: No, no, it's true. But then again, that means people get this experience and you get to question me. Yeah, but it's like a history lesson for you, like school.

>> Steve: I didn't like history.

>> Anouk: Oh, I did. But I think it was because she was lovely looking teacher. Let's. Okay. All right. So, um, what we asked ChatGPT to do was to sit and talk about it, but then she came up with a kind of model that was able to differentiate between the two. And it's a, it's five or six words or statements for each of those eras. Um, and we'll obviously do the earlier ones faster than later, but it's worth comparing the two. So 50s says individual work, paper, secretary, manager and filing cabinet. That is the five words for the 50s.

>> Steve: Yes. I can imagine me something about how it was. So you had, uh, people sitting on a desk working on a lot of papers. Maybe already a typewriter. It was already.

>> Anouk: No, not then. No, I don't think so. There would have been typewriters around, but not.

>> Steve: Not that many.

>> Anouk: They were around the war anyway. But yeah, they weren't necessarily integrated in the business. But yes, they were typewriters. There's no doubt about that.

>> Steve: With a stack of papers on the desk that you can't see, who was sitting behind the desk.

>> Anouk: An in tray and an out tray.

>> Steve: Yeah.

>> Anouk: Ah, managing those papers.

>> Steve: It's maybe what we need to have as well for our admin.

>> Anouk: And I think that. Yes, that's true. Um, I think also I get the vision, the image of somebody sitting on a tall stool and a tall desk overlooking everybody else sitting on low desks, you know, because, um, it was very authoritarian.

>> Steve: A classroom vibe.

>> Anouk: Not necessarily a classroom vibe, stronger than a classroom vibe, but yes, potentially where the teacher was the king of the roost. So potentially that's true. Um, but. But yes. So effectively it's individual work. So people had their own skills. That was also one of the big changes that we see in the 70s and 8. So everybody had their job and they were going to do that job for the rest of their life.

>> Steve: Yeah. And the feeling it gives me with this is that is hierarchy in the company was very important.

>> Anouk: Correct.

>> Steve: Knowing who is, uh, the manager, who is the boss, who do you report to? Knowing the level you are working in.

>> Anouk: True. And we also came out after the war. Of course, that was 10 years before this. Um, and that meant that authoritarian. So, you know, the line of command and the level of the command was the way to go. Um, and so within the company. Yes, you. It was not. I can't imagine it being a pleasant experience. And I can't imagine that the worker of today being able to go back 60 years and actually be comfortable in that environment where they're not allowed to bring their own ideas, they're not allowed to try and improve anything.

>> Steve: Yeah.

>> Anouk: You know, it's difficult.

>> Steve: Won't work.

>> Anouk: No.

>> Steve: Okay, maybe for a few people it will, but, um, not everyone.

>> Anouk: Well, yeah, of course I think that somebody will always like this stuff, but the majority of people know. And also I'M guessing that people didn't move on very quickly because they didn't improve themselves, they just did the same thing day in, day out.

>> Steve: Yeah.

>> Anouk: For 70s.

>> Steve: Yes. 20 years later, um, the keyword stare is employee, department, document, copy, distribute, meeting, decision.

>> Anouk: Yeah. So I think that they learned that the skill set was wider within the organisation and people were not responsible for the end to end process. So there was one person in the 50s or the management that were responsible for everything from one end to the other. No segregation. There would not be a manager in the office, there would not be a manager in the warehouse, there would not be a manager on the production line, there would not be a manager for the drivers. There was one manager that did it all, usually the owner.

>> Steve: Yeah.

>> Anouk: Um, and. Or one manager for a remote location.

>> Steve: Yeah.

>> Anouk: So in the 70s they started sharing

>> Steve: things and work, working closer together in smaller departments, in smaller teams to learn from each other probably and to give better results in what they are doing.

>> Anouk: True for the bigger companies there were still computers. There would be computers around there as well, but most of that would just be for production and um, and creation of, and the operations side of things. So typewriters would still be there in the 60s and 70s. Um, now then, let's have a think about this for a second. So I did start work in the 70s, late 70s I might add, just to be. So everybody doesn't get too upset about all this. So, um, trying to think back to the offices. I remember the first day I started work, which was for what is now British Telecom. Um, we had to create a report but we had to handwrite it.

>> Steve: And

>> Anouk: I can't exactly remember because I wasn't really part of it, seeing like lots of secretaries with typewriters and typing pools. But I can well imagine they existed. But I can also well imagine that. Um, but things like all records for cables and connections and wires, they were all on pieces of cards stuck in drawers. Um, so there was, there was limited computing around in the 70s, that's for certain. But I quite like the idea that you now had departmental leaders and managers. Uh, but everything was still paper based, that is for certain.

>> Steve: Yes.

>> Anouk: And meetings.

>> Steve: Yeah, they started to do meetings and

>> Anouk: agreed decisions, so collaboration kind of started.

>> Steve: Yeah.

>> Anouk: Mhm.

>> Steve: There will be some kind of collaborations in the 50s as well, but it's going to be very, very.

>> Anouk: No, my guess is that unless you were specifically researching it was autocratic. So it was down the line, end of story. Um, there may be, depending on the person, if somebody liked talking to people and sharing Ideas, it might happen. But I can imagine in 70s it was all about. People were saying the way to get things done quickly is a fixed process. And you, you one person will manage that. So very rarely would you get that. I know that's what you think because that's all you've lived with.

>> Steve: Yeah, true.

>> Anouk: Didn't work that way. You basically, you know, if I've got to deliver these 10 things to these pubs, you would be given a list and that's what you would be expected to do. And if you deviated from it, you'd be fired, end of story. Because they have no way of fixing this or people didn't have their own creativity. So people change. It's interesting. It's a People change.

>> Steve: Yes.

>> Anouk: All right, so that's 20 years. So in 20 years, that was quite a long amount of time. In 20 years you went from, uh, individual work to department to work.

>> Steve: 20 years.

>> Anouk: All right.

>> Steve: It's a long time to change things like that.

>> Anouk: Yes. So the 90s then. Of course we know there are computers around then. So we have employees and computers shared

>> Steve: drives, email to colleagues, meetings and decisions.

>> Anouk: I guess we did have. Yeah, but they weren't necessarily cloud. Yeah, so let's do that again because I disrupted everybody. I'm sorry.

>> Steve: So in the 90s we had employees, computers, shared drive, email to colleagues, meetings and decisions.

>> Anouk: Yes. In the 90s that's when we shot ourselves in the foot with email.

>> Steve: Yeah. It is time that the Internet arrived and that computers were more often used in, uh, companies.

>> Anouk: The Internet was not really there then. It was slowly arriving, but not in any way that meant everybody had an Internet connection at home. No, there was networks and email was delivered across business.

>> Steve: More in business related work. Emails were arriving, Internet was arriving. It's the time that I, um, learned to type at school in the 90s and I did learn it on a typewriter and not on a computer.

>> Anouk: Yeah.

>> Steve: So I know the feeling of a typewriter and if you misspelt that, you need to rip your paper and start over.

>> Anouk: Did they not have, um, what was the white marker stuff?

>> Steve: Um, yeah, you could use that as well. But that was not allowed with us. You need to restart over.

>> Anouk: Not part of your test. I had to get out of it. It's a little bit like when I was at school and calculators came out. Why do we need to do maths now we have calcul. So typing. Well, because you need to learn how to do it properly. Typing. But I've now got markup. What's it called, uh, Shit, can't remember. But no, you're not allowed to use that because you're supposed to learn Tippex. Yes. You know, learn how to type.

>> Steve: So what I feel with this evolution is in the 70s we had employees and departments working together. Now you in the 90s, you have your employees, your computer and your shared drive. You still work together, but it's a, ah, less personal connection than in the 70s.

>> Anouk: I think the biggest difference if we take those three is that you have an individual doing a job and they did that job for the rest of their life. Uh, in the 50s you had that person working with other people in the department and they shared the work. In the 70s, yes, sorry, 70s. And then in the 90s you had those departments that interacted and shared their work. So it's really the breadth of computing that changed. Sorry, the breadth of communication that changed. Um, and again, so now we're talking 40 years just to get from individual workers to the point where some computing was done at a operational level. And uh, I'm guessing there were laptops around at that point, but not many. Well, there were laptops. It's debatable whether you call a computer the size of a briefcase a laptop, but yes, there were laptops and desktops around. Um, very little sharing of content, even though they had shared drive. Um, because people really didn't understand it.

>> Steve: No, people didn't like to share the work they did. It was still very new and still afraid that something, somebody would do something wrong with it.

>> Anouk: Correct.

>> Steve: Trust was an issue, but that still lives today.

>> Anouk: True. Everybody puts things on their OneDrive and doesn't share it with anybody else. Yeah. So we failed in the change, really. So when we have change forced upon us a little bit like we've seen in that 50s, 70s and 90s, then they put up with it. But if it's personal, they still have a, you know, a penchant to keep control of it, I guess.

>> Steve: Yeah.

>> Anouk: Um, I don't know whether I do that. I do, I care. The thing is that I think in the 50s a document that was typed up or, uh, and then distributed as a, uh. This is the way you will do this job. That was then put into a filing cabinet and it was then kept until it was updated again, which in many cases would only be 10 years later.

>> Steve: Yeah.

>> Anouk: And so that piece of paper had a longer period of life, whereas. Because we can now change that piece of paper quite easily, your average lifetime of a Word document or a spreadsheet you create is months.

>> Steve: Yep.

>> Anouk: Three months, four months. So therefore sharing it and keeping it and stuff is.

>> Steve: Yeah. I was providing a training yesterday morning and there was a very interesting discussion going on. Um, so a guy in my training class, he created a document, he saved it on his OneDrive and um, after that, when he was ready to share it, he copied it into a Microsoft Team, but he kept the local version on his OneDrive just in case somebody would change something in the documents.

>> Anouk: Right.

>> Steve: So that he can always restore it and back to the original. And then I've told them about versioning and all of that and then he was like, that's much easier.

>> Anouk: But why would he want to make sure nobody changed it?

>> Steve: Don't know. That was not clear from the entire discussion. I said, isn't that working together on documentation on what you guys are doing? But no.

>> Anouk: Oh well,

>> Steve: yeah.

>> Anouk: And I think you still get a lot of that happening. I think some people are a bit. I think some people are still put into difficult positions by people that don't know how to manage people properly and all that kind of stuff and uh, get the most out of people. So that is a bit of a thing. So 40 years of change. Not a lot.

>> Steve: No. If we jump to now, that's an interesting one.

>> Anouk: It is.

>> Steve: And recognisable for everybody of us. So now you have your employee, they have a laptop, they have 37 tabs open in a browser, they have endless team Spins, they have 12 meetings to attend and then they trying to make a decision.

>> Anouk: Wow.

>> Steve: The evolution you see is the making of the decision is being much more difficult nowadays than you had in the past because everybody wants to be involved and wants to get in, have to say what they want to say and be involved in making that decision. Why? There are too many people trying to make something.

>> Anouk: That's true, isn't it? So in the 1950s, one person made the decision and they were responsible for it, generally accepted the responsibility and away you went. So nowadays people tend not to. Some people don't like making decisions. So that's really a distinction of leadership now, making the call and accountability. So we understand it a bit more. So in the 70s, decisions were made by a smaller group of people. But as we've added more increased capability, making a decision or getting a decision made is harder.

>> Steve: Yeah.

>> Anouk: Interesting. And of course now we also have AI to also confuse the decision making process.

>> Steve: Yeah.

>> Anouk: I wonder whether anybody's been fired yet for making the wrong decision and then saying, but AI made the decision for me. I just did what I was told.

>> Steve: I Don't think we are at that level with AI already.

>> Anouk: Um, really? So you don't think agents are uh, automating stuff and making calls?

>> Steve: They do, but not the decisions. And uh, not that people will get fired about it. Not yet.

>> Anouk: I was being facetious. If anybody decides can't stand up to a decision they wrote down from A.I. that's not true. But, but I disagree with you. I mean I think that AI is working in banks and investment houses and making those decisions all autonomously on their own. And I think more production systems are making those decisions on their own. And I think in, in I, I, I, for some reason I joined a call centre distribution website just recently. Uh, their newsletter keeps coming on. Um, and they are also starting to talk about how AI is now starting to make decisions on call centres and customers and things coming in and emails. Um, because I think the real problem is that today we accept more numbers of failure so we accept less quality for lower cost. And AI can drive that. It can make mistakes. But if it makes 5% mistakes and that doesn't cost us 20% of our budget, then that's acceptable. But we just live with those unhappy people because some people are going to be unhappy anyway. From call centre.

>> Steve: Yes, of course.

>> Anouk: So I don't know. I don't think there's a gap. So there's interactive AI at the bottom where a human is involved in driving it and making decisions from it and then there's automation at the top where we're using it to streamline existing processes.

>> Steve: Yeah, with the ones you mentioned about in bankings and in banks and all of that. I still think that there will be a human in the loop to review all of the decisions.

>> Anouk: Say there wasn't. No, I don't think they'd review them, but they would be, I think they would be allowed to make the call. You have no choice. Because if one bank is using uh, an AI to manage stocks and shares, the other banks have to do the same.

>> Steve: True, but I, I still, and that's my, my way of thinking and my opinion is that there is a human that is saying this is the right call or not. Nope.

>> Anouk: I don't, don't believe it.

>> Steve: We don't know.

>> Anouk: I do know, I uh, do know that there are systems that are automatically buying and selling shares for banks. Now they will probably have some strong guide rails and there may well be that, um, if there's certain characteristics, humans will be in the loop. Okay. You're building workflows for people that are fully automating. Things and people are not in the loop. That may or may not be using AI, but what you write in PowerApps you can do in AI. So true. So the whole point of between 90s and now is that we have removed decision making where it's a simple automated decision.

>> Steve: Not yet, we didn't. Um, we are removing it. It's not yet removed. A lot of people are still in the loop to make the decisions.

>> Anouk: No, I really, really disagree. A production factory will have automated ordering and delivery of stocks based upon things where people order a new car. So if you order a new car, I know this for certain for the last 30 years, if you order a new car, all the parts that make that car, the 50,000, they will be automatically ordered and a delivery date provided with no human in the loop.

>> Steve: But that's not a decision making. That's ordering something after a decision has been made by a human. A human person orders that car, he decides what options he wants, which, which colour it needs to be, which colour of sofas it needs to be. And then the machine is ordering all of the pieces to build the decision of the process. Yes.

>> Anouk: Okay, so whereas you think that, um, somebody will decide which stocks and shares we're going to buy and sell and a computer will buy and sell them.

>> Steve: Yes.

>> Anouk: Bullshit. It's got to be. There's going to be a computer monitoring the price of shares. And um, basically based upon some kind of model, there'll be decision that says get in there now and buy those shares if this and this and this happens. And there's also AI sitting monitoring and saying, you know, um, um, these shares here have gone up and they've gone down and they've gone up again. And if I see that up and down and up again, it usually means that there'll be a sell. So if I buy some now, I know that people are going to move in and then I can sell them at price and I'm fairly certain within a budget, maybe 10,000 or something. All right, maybe even half a million, depending on the size of things that they're dealing with, they will be able to make those automated decisions themselves.

>> Steve: I would think that's possible, but I think we're not there yet. But that's be allowed to disagree.

>> Anouk: Uh, we are. People can tell us. But yes, I'm absolutely certain that is the case. I know there's been some experiments done where people started off with a hundred dollars and they say you got a month to turn it into more. Go away and do what you want to do without any human involvement.

>> Steve: But I'm very sure of what you think that we already have. Five, ten years, times max. It will be there. It will be more automated than what it is now.

>> Anouk: Really? Yes, obviously it will be more automated

>> Steve: than it is now.

>> Anouk: But you think so? Let's go back to the question we asked AI and we'll come back to this in a minute. So one of the questions I asked when we did this was, okay, if we was to estimate in some kind of model what percentage is now human decision or. No, um, basically I asked the question, if we go to the 50s, what is a decision? How much of is a human in the decision loop? And it was 90% and 10% is, you know, done. Whereas today, of course, it's probably 70% is automated with the humans making 30% of the decisions. And then, of course, the future, it will be more automated and then we'll end up with superhumans. So we know we're going to move on. Whether I'm right or wrong about ordering stocks, it's going to change. The real question that I wanted to ask is, are humans more intelligent? Intelligent may be the wrong definition. Capable, responsible, accountable, better decision makers than they were in the 50s because of the way that the technology is driven or the way that business has changed. Or were you not listening to the question because you were trying to find out what we put up on AI?

>> Steve: No, I was listening to the question. Humans are not better decision makers, but I think humans are making the decision more about, um, the past they have carried with them, um, and what they have seen and where they went wrong than AI. AI doesn't have that post of what you have been seeing, what you have been doing, what was a good decision, a bad decision about you. So it's been, it's different.

>> Anouk: Okay, so my question wasn't about what AI is doing. My question is whether the humans are better decision makers. Yes, but since the 50s and 70s and 90s, not with AI. So were human beings better decision makers or entrepreneurs or, uh, accountability in the 90s than they were in the 50s? And is it because of technology that's given them that? So people in the 50s did not need to make a decision. They followed step one, step two, step three, step four. Okay, in the 70s they were dragged in a departmental meeting and the department may well get the head of the department says, hey guys, we need to, um, try and reduce the number of, um, um, widgets that we use to create our product. And so they then talk about and work out how that would happen or we need to reduce the price so they have uh, more meetings about some of those decisions. In the 90s of course you had entrepreneurs and dot coms and startups, so I'm guessing there's less structure then humans took that opportunity to actually spend more time thinking and improving and becoming potentially more intelligent, more assertive, more driven. So what is going to happen now? If you've got AI, where are we at? But that took 40 years to do. We've only had AI for five years, which is why. And I, so I do wonder whether we're actually changing as quick as everybody seems to think they are. I know we are changing. So some of our industry, software creating and all that kind of stuff, it's a different ball game.

>> Steve: Yeah, it is changing and not only in our industry but also in personal lives because a lot of people are using AI nowadays for a lot of things. Trying to understand. And um, like I did uh, recently I checked with my bank accounts to see how, which hidden subscriptions I had that I didn't know to cancel them.

>> Anouk: But you could have done that every month just by having a regular routine check?

>> Steve: Yes, but it would take me much longer than I have done it now.

>> Anouk: So because it was easy, you did it because you didn't have to go through with a pencil check in every box and stamping. So because process is easier, we have a different thought process?

>> Steve: I think so yes.

>> Anouk: That's probably a reasonable way forward. I mean I know a few people that have done that and I keep saying I must do that but I haven't got around to it. Uh, but I find it quite interesting that not everybody does it or even that it's not automated. Why haven't the banks given you that facility?

>> Steve: I'm um, guessing very sure it's because uh, the law doesn't allow a bank to give you AI to your accounts.

>> Anouk: Why?

>> Steve: I think you need to ask the bank for having it. It's personal data that you will share with an AI system. So they need to be sure that it's secured.

>> Anouk: What if the AI is running in their data centre? It's going to be as secure as it is. Now AI is no different to a database query, true in terms of what it's doing. So as long as it's secured in that area. But yes, I can imagine that banks would be nervous of what AI might uh, do and whether if somebody gets to that AI.

>> Steve: But, but they maybe are developing this and testing it on so that we don't know.

>> Anouk: Well it doesn't need AI to do this though, does it? So you, they could just use a query that says hey, you know, give me and some banks like I think we both have the uh, Revolut account tells you how many what you spent shopping and, and restaurants and everything else. So they build those features in play. But yes, when you get an MCP for your data service it would be easier because I hate. All you can do at the moment on my bank system is filter by date or by value and it tells you what's on your statements. I mean that's it.

>> Steve: Mhm.

>> Anouk: Yes, but maybe that's the way it should be. Maybe you download it and then you can feed it into AI and sit there and say okay, give me a rundown, tell me what's unusual here. Yeah, I used it to do my contract, you know, I want you to go through this contract, tell me what I won't like. Um, and then that meant that when I read it I didn't have to focus every paragraph because I'd already got the highlights. So as long as I sort of got a feel for it. So I trusted AI to sit there and identify them. I've signed it, now it's too late. Yes, if AI me, then you trusted

>> Steve: AI to identify it, then you read through it yourself and you made the changes, you made that call.

>> Anouk: And then when I got the new one back, I showed up the original one and the new one and said okay, just tell me the differences. And then I uploaded the one where my changes went in. One change I need to make on this mixing desk software, uh, on my new laptop is to turn off the go to sleep when a particular wonder whether it will allow me to do it when a particular application is running. Good question. Well anyway, it's interesting to look at that 50s, 60s, 70s, 80s and 90s. And the real question is of course where is it going to go next? Because these changes have been 20 years in their making. Yeah. So from 50s to 70s we went from individual people just doing what they were told to do to actually departments deciding what they needed to do. All right, with more leaders that had uh, to talk to each other above the department level to work out where they wanted to go. Which is great. That uh, took 20 years. We're only five years into AI.

>> Steve: Yes.

>> Anouk: So what do we think? And what did our AI think the future would look like other than the sarcastic comments she gave us?

>> Steve: She thinks it's going to be something like employee, AI agent, virtual workspace, augmented reality, augmented reality meeting, automated report and human Wondering what is real.

>> Anouk: Humans wondering what is real. It's a bit like that already, isn't it?

>> Steve: Yeah, but I can see the first bit. Employee AI agents that you get an AI agent as your closest employee that is challenging you about what you want and where you need to go to and how to go there.

>> Anouk: Yeah, I think so. I mean, um, um, there are, uh. I was talking to you about this the other day. Um, when was the last time you actually opened Word to create a Word document and start typing it?

>> Steve: Um, no, you were not talking with me that till last day, but for me it was actually on Monday.

>> Anouk: You actually started off with a blank sheet of paper and you didn't use AI to fill it out.

>> Steve: I did. I didn't use AI to fill it out, I used AI to review it afterwards.

>> Anouk: Interesting. I can't remember the last time I actually opened a blank Word document in the last year. And even emails I do. I have to say if I'm personalising an email, then I would write an email and I ignore the copilot prompt that says tell me what you want. And there are some real frustrations. I think you watched me a few days ago screaming at, ah, Holly because even though I told her I wanted something done, she just did not get it done and then forgot things.

>> Steve: Lightweight. That's what he's saying now. It was not screaming, it was everything else.

>> Anouk: I was ready to throw that fucking computer across the room.

>> Steve: Yes.

>> Anouk: So, uh, so yes, you need to be. There are some frustrations around it, but you know, I. So as part of something called Frontier Crew, um, I'm developing something for AI as a service where companies can just buy the service and they don't have to think about tokens and they don't have to think about installing it and they don't need to think about developing it. Uh, it's just going to work. The only thing they need to do is to edit content and um, put their relevant things in there. So I was creating policies for HR and policies for finance, which are sizable amount of documents to do. Um, and the idea is that you can then turn off on and off certain policies and your AI will behave. Um, but creating those documents was interesting because I wanted. So I was creating 25 page documents, uh, and I wanted the titles and style 1 and style 2 and style 3 all separated and updated and you have to tell AI exactly what you need and if you forget to tell them that, uh, I think I said to you that it's like working with an intern, you know, you Kind of got to say, look, can you please give me a one pager that says this? And then you have to tell them what they need to do so that they. Because they've never done that kind of stuff before.

>> Steve: Yeah.

>> Anouk: And it is an interesting process. So I think that we still got some frustrations to go through. Um, a long way to go. But, um, yeah, working with agents and AI, I really don't know when we will not have Word documents and Excel documents and PowerPoints. You don't write PowerPoint anymore?

>> Steve: No.

>> Anouk: You. Even if you do it in your graphics software, you use AI?

>> Steve: Not always.

>> Anouk: Not always, no.

>> Steve: Ah. I did create one from scratch completely myself.

>> Anouk: Just for old time's sake.

>> Steve: Yes. Because I had the time to do it and it's fun doing. Ah.

>> Anouk: Uh, yeah, that is interesting. And I think it's a little bit like drinking a whiskey. You know, sometimes you can take a whiskey and all you need is to have that shot and that burn and away you go. And other times you want to just sit there and relax and enjoy it and learn from it.

>> Steve: Yeah. But the one that I created from scratch myself, again, I tried it with AI and it was not giving me what I needed.

>> Anouk: Yeah. We still got a long way to go. So even though AI is supposed to be moving rapidly, uh, it has made some significant changes where the data is available. So like software coding. There's so much software code around, it can very rapidly, through machine learning, know what works and what doesn't work and everything else.

>> Steve: Yep.

>> Anouk: And API commands, same thing. It can do that very, very well. But that's quite industry specific.

>> Steve: It is what I now see. Um, I provide a lot of trainings and all of that and I get more and more questions about. All right, we want to get involved, we want to start with AI. What do we need to do? Um, the training I did yesterday, they started, uh, pilots, um, for Copilot. So in a company, about 200 people working, five people now have a co pilot licence to see what it does and how it can help them. Them.

>> Anouk: It's the wrong way to do it, isn't it, really? I mean, if a company doesn't know, uh, I think that's potentially. That's an interesting thought. Give me a second.

>> Steve: Yeah.

>> Anouk: So if we think 50s, 70s, 90s, it started off with fairly strict processes.

>> Steve: Yep.

>> Anouk: And then it moved into less dependency on the process and even less dependency on the process and more automated thinking, mainly because industry's got bigger. And then they were quite fluid because they looked at what they're doing with AI because we think that it's magic or because a lot of people think that it's magic, they still need to think about those processes that they want to improve.

>> Steve: Yes.

>> Anouk: And so the way forward with AI is okay, not how can it change? What can it change? What is expensive to do that? We can potentially investigate AI doing it. What is, uh, expensive in two sides? 100 people have to spend 10 minutes every hour doing this. Or one person has to spend 60 days doing this. You know, there's the value. So if you can replace those things with AI then you're going to get some value out of it. And it is about the value line.

>> Steve: It is.

>> Anouk: And that, in theory is what Frontier Crew is supposed to do.

>> Steve: Yeah.

>> Anouk: This AI as a service is supposed to sit there and help you ask the questions and do the things that from a finance perspective, for example, I need, or from a HR perspective that I need. And they're about processes, not about replacing people. It's about making the processes easier and simpler. So that's a good question. So if we really want to make the change, we should be asking about what can I change? Or what costs me a lot of money to do? And then how can AI make that saving? Yep.

>> Steve: Uh,

>> Anouk: I wonder whether this is typical of the times that we've been through in the last 20 years with social networking and communicating so much stuff so easily. Everybody gets to see the words AI Yes. My daughter doesn't believe any of them. Of course. Anna, uh, get real. AI is here to exist. You can't keep ignoring it. But that's because her background is literature and books and words. So, you know.

>> Steve: Yes. I think the industry you are in, we are in is very important and we embrace AI because can you imagine, uh, all of the work we are doing now that you don't have AI to do it?

>> Anouk: Yeah.

>> Steve: Creating all of those documents for Frontier Crew without AI that will take M days.

>> Anouk: I would have had to get somebody from Fiverr to do it. I would have had to put humans in the. Yeah, cheapest human labour to write up.

>> Steve: But I do understand there are industries where AI is not that big and not that heavy involved already. And that they think like, all right, we don't need it.

>> Anouk: Yeah, it's true.

>> Steve: And of course everything you see on social media, um, is now questionable. Is it real or is it A.I.

>> Anouk: yeah, that's true. We know we've had a groundbreaking, um, legal precedent in Germany, uh, this week, um, is, uh, around, um, uh, the fact that these generative AI tools have used music and stuff to create their models. But the models guidelines were not restricted enough to say don't reproduce the same song. And they've now passed through law that that was basically recreated the same song that I created 20 years ago. So. But I guess it's normal, quite honestly. I mean. And you get it in, uh, in the book industry. You get it in movies as well. I mean there are some movies. Shawshank Redemption is an amazing film, all right. And it was up for seven Oscars until somebody said, but just a minute. I wrote that storyline five years ago, proved that somebody somewhere else had seen it. And then basically you end up in a court case because even though the film is now produced and done, somebody said, but I wrote that storyline. Somebody stole it.

>> Steve: Yeah.

>> Anouk: So it's not that it was new, it's just that we can do things quicker and match these things up.

>> Steve: Yes.

>> Anouk: So it's interesting times ahead.

>> Steve: It will be a nice way to see how everything is evolving.

>> Anouk: Well, I think one thing's for certain. When we talked about the 50s, we talked about 90% process, um, 90, uh, percent human and 10% process. And so. And. But now we're probably at, ah, I think 70% process and 30% human, but more intelligent human, more adaptable human, more accountable human, um, more entrepreneurial human. To be able to get the best out of that 70%. And I think that's going to change even less. Um, and I wonder whether we'll end up going back to specialists, people that are specialists, uh, in an area and they're staying there longer because that area develops quicker using AI and improves quicker using AI. The future will tell us the future is AI. But we knew that anyway.

>> Steve: We knew.

>> Anouk: Actually, no. The future is superhuman.

>> Steve: Yes. Maybe.