Navigating the Future: AI, Disruption, and the Rise of AI Agents

Speaker A: Foreign.

Speaker B: Welcome to the Unrivitive podcast, where we talk about digital transformation, artificial intelligence and people. Today's podcast is brought to you in part by our book, AI in a Weekend An Executive's Guide. Please click and buy and make us very wealthy. John, with that, would you introduce our guest?

Speaker A: All right, thank you, Martin, always the, uh, pitch man. Uh, so thanks for the um, info on the book. Uh, today we are joined by Dr. Mark van Rymanam, uh, who is a strategic futurist and the digital speaker. Uh, so Mark is joining us today from uh, Australia, I believe. And thanks, uh, for being on the show, Mark, uh, as we typically do, you know, we'd like to turn it over to you as our guest, uh, and let you give us a little quick, uh, background on yourself and how you ended up where you are, and then we'll dive into the conversation. Uh, so I'll turn it over to you, Mark. Thanks for being here.

Speaker C: Well, thank you for having us, Martin and John, it's an absolute pleasure to be on the show. So, uh, Yes, I am Dr. Mark Verebenam. I'm a strategic futurist, which, uh, means that I think, uh, about the future from a technology perspective. So I look at how, um, new technologies such as AI Quantum, um, um, the Metaverse, um, and many other technologies are changing our society, are having an effect on organizations around the world. I'm, um, m a keynote speaker, so I travel the world. I'm fortunate enough to visit, uh, all continents and many countries on a regular basis, um, to help organizations and governments sort of understand what's happening within this space. Um, and at the same time I really like to practice what I preach to be able to understand what are sort of the implications of these technologies, how they will have a good, a bad and an ugly, um, effect on what we do, how we live, um, how we work. Um, so that's why I created the brand, the Digital Speaker. Um, so I am present in the digital void. Uh, so I have a, uh, digital twin that actually speaks, um, 29 different languages. And you can talk to me via text, audio and video. The first person in the world to do that. I, um, delivered the first TED Talk in virtual reality, uh, uh, four years ago, which is quite exciting. Um, and I'm currently building a new, uh, media platform called futurewise, where we help you, um, move from chaos to clarity in this deluge of information and misinformation, um, and this vast disruption that we see in society. And we help you understand what's happening using hyper personalized Insights. So, ah, that keeps me busy. I've written a few books and I'm currently working on my sixth book.

Speaker B: Marc, thank you. Thank you so much for joining us. Thinking about um, some of the areas that you probably think about daily, like the future landscape. Here we are in 2025. What do you believe will be really critical for enterprise companies to be moving on now? And what can we expect, uh, when we link AI to their future?

Speaker C: Well, that's a very good, uh, good question to start with because there's so much disruption happening at this moment. It's incredible. Now everyone talks about artificial intelligence at the moment and that's of course sort of the foundation of all the disruption that's happening because it's sort of powers all the other technologies. But there's so many technologies converging from robotics to quantum computing, spatial intelligence, um, uh, 3D printing, uh, uh, biology, um, so there's so many technologies that are having an impact on the enterprise that it becomes really crucial for organizations to become aware of what's happening, to sort of start to uncover m, the signals that are out there to understand where the future is heading for their specific industry, for their specific organization. I, uh, think that's sort of the most crucial component that organizations need to start doing is understanding what's happening. Um, and then uh, from understanding what's happening you can then start to adapt your organization and incorporate these technologies, such as the upcoming AI agents, which are supposed to really have a big impact in 2025, um, and really change how we work, um, the work that we do, um, and it will have a big impact on the economic activity, um, of organizations. But I think with all that disruption it is really important to be able to verify what's happening, you know, because there's so much, as I said already misinformation happening. Just today Facebook announced that they have killed their fact checking, um, uh, organization within Facebook. So very problematic I think. Um, so misinformation is crucial. Um, and you know, from there on, eventually we need to start empowering everyone, all stakeholders to embrace these technologies and prepare for what's happening. Because seriously, if you're not paying attention and empowering your stakeholders, you will be left behind in the years to come.

Speaker B: Oh yes, um, let me unpack one more and then I'll hand over to John, uh, and then we'll go back and forth here. What do you think are some of the biggest hurdles for many of the multinational enterprises to be cognizant, considerate, um, about AI and an impact, um, and what May they need to do to overcome this.

Speaker C: Well, I think it all starts with education, um, uh, especially in a world that changes so rapidly, where yesterday invention is, tomorrow is tomorrow's mass product. Um, and um, if you're not paying attention, if you snooze, you lose. Um, so within every, at every level within your organization, everyone needs to be aware of what's happening. How is AI changing? Um, how is AI having an impact on the work that we do within our organization? How is it going to have an impact on our industry? Um, what can we learn from that? How can we embrace this in a way that is beneficial, uh, to our organization, but also to the wider stakeholders, as everything is interconnected in this world. So I think that's really, really crucial that as organization you educate all your stakeholders, which means your first and foremost, your employees, um, and your leadership team of what's happening, but also your customers, also your shareholders. How is this going to impact, um, the business? And once you have a good understanding of what's happening, then you can start moving forward. But a lot of organizations, um, sort of lack that understanding, uh, because they're just sort of a fight or flight or fight M mode we are in because there's so much disruption happening that nobody really knows what to do anymore. And I think that's sort of the challenge for 2025.

Speaker A: Totally agree with that. You know, it's, it's a balance between the hard technical skills and the soft skills. And I think typically the technical skills are what receive the lion's share of attention when uh, people think of AI because it's just inherently a uh, technical topic. And um, when I started out in the field in you know, machine learning, uh, it was a very gradual learning curve for me. And then I would say sometime around the end of maybe 2022, uh, when people started using chat GPT and then all through, um, you know, 2023 and 2024, um, just trying to keep up, trying to keep pace with everything changing. And all these new models and all these new developments and applications became so much more uh, an exponential growth compared to when I was learning machine learning, um, which obviously is part of AI, but um, big change, uh, and uh, very hard to manage. So, um, thinking again, you know, about AI and maybe getting a little bit more specific, uh, in 2025, are there any particular, uh, or maybe impactful applications that you see at the enterprise level for using AI and that can again, you know, be from like a purely technical standpoint or even from like a, uh, people management, uh, which could be like customer experience. Um, uh, what are your thoughts on that?

Speaker C: Well, I think 2025 we'll really see the transition of, from generative AI that allows to create content, to create images, text, um, even videos, or even entire, uh, immersive virtual worlds to AI agents, um, uh, agents that can work together with each other, that can, um, solve complex problems, um, and can really take over major parts of the job that we humans normally do. Um, and that's going to be very, very transformative. Um, maybe not so much in this year, uh, because it takes time for these AI agents to become better, um, for organizations to play with them, to, to experiment with them, to integrate them in the work processes. But certainly the years afterwards we will see these AI agents becoming a lot more advanced, a lot better at collaborating with each other, a lot better at managing itself, um, um, and as such, um, become better able at solving complex problems. At the same time, I think the more we humans are exposed to these AI agents, we will also become better at managing that relationship, at managing the relationship with AI, at understanding what our role as humans is when we talk about, um, integrating AI agents in the workforce. Because I do believe that there's fortunately still a role for us humans, uh, within organizations. And uh, we are the ones who should manage these AI agents and not the other way around. So, short answer, AI agents, um, uh, is definitely, uh, going to be the big thing this year. M. And they will be integrated, um, in the coming years, um, everywhere, at every part of every organization.

Speaker A: Makes sense. Yeah. Agentic AI has been a main focus, uh, for my research and interests as well. And that's all people are Talking about in 2025 is moving away from the novel, you know, kind of proof of concept use cases that AI was being used for maybe through 2024 and into. How do I actually add value to my organization using these tools, um, and move into incorporating them in, you know, workflows that currently exist.

Speaker C: Uh, which is by the way, fascinating if you look at this because only two years ago we had the introduction of ChatGPT. That's only two years ago, November 2022. Um, and the amount of progress that we have seen in the past two years is just absolutely mind blowing. Um, and um, that's only going to speed up because, you know, as I like to say, we have sort of, we have, we've crossed into the second half of the chessboard. So now we are at exponential territory. So everything is moving faster and faster and everything is going to move faster and faster sort of A compounding interest of ah, of, of innovation that will completely change everything that we do.

Speaker B: I, I like where how you describe we're, we're on, we're at the other side of a chessboard. Um, um, I like to also lay out that I think this chessboard is multi dimensional and it has you know, three levels at least or more. It's not limited by the levels and we're, we're just tapping second level and third level, uh, application of this and that includes you know, mining your data and cleansing your data. And you know, along that vein, um, one of the areas where it might become interesting is the role of synthetic data and how that could be used at an enterprise level, uh, for great advantage. Have you delved into that area and considered what are some of the impacts we might need to think about?

Speaker C: Well I think you strike a good point because data in the end is the absolute foundation of all these things that are happening at the moment. Because without the data we can't do anything. Um, without the data we can't understand our customers, we can't um, uh, improve our processes. So we need to have the data and we need to have high quality data that we can actually trust and that we can verify, that we can use. So data is really the foundation of everything that we do. Um, now your comment regarding synthetic data is really interesting because uh, a lot of the data that we've been using and from especially public data on the Internet, we sort of reaching the end of what's available because uh, even though we have tremendous amounts of data to our disposal, even that's not enough if you want uh, to train this data, uh, to train these algorithms. So we need to move to synthetic data but we need to be able to ensure that we can trust the synthetic data that we are creating. Uh, the advantage of synthetic data is that we can create the data in such a way uh, that it is high quality, that it is trustworthy. We need to be very sure that when we do that, that we have the verification processes in place to make sure that the synthetic data that we have created, that we are using to train our models is indee reliable, unbiased, high quality. Uh, so for the past decade um, their focus has been um, on data and I think that's only going to become more and more important. That component is not going to go away. In fact, I think it will only increase in importance the more we integrate ah, these technologies into our organization, um, to ensure that as humans we stay in the loop and as humans we have Control over the situation.

Speaker B: Valid point.

Speaker A: It's very true. I think, I think our first podcast that we did, Martin was start with the data, wasn't it? Wasn't that the topic? Um, yeah, and I don't think that's changed. You know, we, this was way before uh, Chad GPT, I think was even a ah, thing. So, so yeah, you're, you're, you're, you're right on point there, Mark, that uh, the data is extremely important. Right, yeah.

Speaker B: So just, just to build on that John, I mean the basis of why you would use synthetic data may be not transparent for some of our listeners. Some of the reasons you may use synthetic data is to help balance out bias for gender, balance out bias for ethnicity or social economic class data. Also to build out synthetic data sets for testing which is actually really powerful as opposed to private, you know, personally identifiable information for individuals. So synthetic data can replace personally identifiable information. That is uh, you know, of concern. So that's one of the huge upsides of synthetic data.

Speaker C: Absolutely. And I think that's, that's important to be aware of that, that you know there is um, but that's, I guess that's whatever technology there's, you know, there are very strong upsides but also risks that come with the data, with the technology. Um, and that's also with synthetic data. We need to make sure that you know, the data that we use, uh, to fill the gaps that we have when it comes to buyers, when it comes to geographies, when it comes to target groups that ah, we make sure that this of high quality and we don't uh, make the biases worse from the perspective.

Speaker A: Very true, very true. And I've always had this hesitancy thinking about we have AI training, AI now in this potential infinite loop, but the human in the loop, the human um, element is still very much a part of that because of the trustworthiness. Um, so I don't see that that's going to be changing anytime soon. I hope not.

Speaker C: Because otherwise if you know, if the machines take over and they ditch us, I think we have a bit of a, bit of a challenge.

Speaker A: Could be a problem. Could be a problem. Um, let's move forward and uh, get your uh, futurist uh, brain thinking here which uh, I guess maybe we've already been kind of talking about uh, up to this point but thinking about trends uh, in AI and I know we've talked about agentic AI, synthetic data, but is there anything that you see, um, or maybe you know more than Martin and I About uh, what's happening in 2025 and beyond. But are there any emerging uh, trends that you see that people might not be talking about in AI or even maybe you know, things that you would uh, anticipate um, happening within 2025 and beyond, uh, that might not currently be in the um, you know, the news cycle, so to speak?

Speaker C: Yeah. So ah, what I find quite interesting is how um, AI is becoming better at the reasoning component, um, and thinking through um, different processes. Um, as an example I'm currently using and playing with ChatGPT Pro, um, so that's the $200 a month subscription. Um, and uh, the way it is reasoning and thinking through the steps and processes, even by giving a very simple prompt, um, is quite fascinating. Um, so you can see we're definitely making a big leap there. This is still the O1 version, not even the O3 version, which um, Sam Alvin announced like a couple of weeks ago. So it's clearly to see that AI is getting better at reasoning. So we're moving away from some sort of being the parro, um, where it just, you know, just mimics the data that it has, that it has given, um, um, and sort of, it's sort of a mirror of our collective human conscious to thinking more, uh, and reasoning more about the question that it has been asked. And what I find fascinating to see is that quite a few times now um, I have. AI have given me like a completely novel approach to something that I was thinking about, um, which I think is fascinating and stuff that I am actually incorporating in the work that I do because of that. So um, I foresee that reasoning capability will improve in 2025 and with that it will become more useful also in academic research in innovation, um, where we can uh, move faster, uh, as we develop and invent and innovate uh, in a variety of domains. So I'm very um, excited about that particular component of when AI moves away from being a parrot, um, to really thinking individually, um, um, at a completely different level than we humans think of course. Um, ah, but to have a different approach um, and linked to that is also where uh, we move away from the token approach where we sort of predict the next word to a more like a concept approach where we think in broader terms like sentences or paragraphs, um, and how they link with each other to get better insights. So those two components I think are, they are related to each other and they are very fascinating to see how AI will evolve um, in the next year. And yeah, the change that we'll see in the uh, coming year will be quite large and quite significant. Uh, I think that's an understatement. Um, uh, and I'm really excited about that.

Speaker B: Excellent. Um, are there any companies on your radar screen to make aware that we should be watching for 2025?

Speaker C: Well there's so many new companies that appear on a daily basis, you know, it's impossible to keep up. Um, apart from the big five, the big ten, uh, AI companies that are, that are pushing really, really hard. Um, and a lot of the startups that come that follow the big five, the Big ten, um, use the technology of the big ten. Um, so it's, you know, we all, everyone is benefiting. Um, I think when it comes to AI, the good thing and the problem here is that the size of the organization and the resources that you have as an organization are directly linked to the progress that you can make. Um, because um, these developments require so many resources from human uh, to capital, human um, capital to actual capital. Um, that is, that's why these large companies, anthropic, OpenAI, uh, uh, meta Google, um, or the Chinese counterparts, um, can move so fast at the moment because they have access to these resources. But that's also really problematic because we get a very centralized, very tiny elite of companies who build all these powerful technologies and the rest just has to follow. And that's just can only leverage the technology that these big companies are developing. There's no point in naming specific companies, um, uh, because they might appear new ones tomorrow or disappear the day after, uh, because everything is moving so fast.

Speaker B: I was looking for our stock. Tip of the day.

Speaker C: Sorry, no financial advice here.

Speaker A: Well, I think maybe the last question, um, to wrap things up is thinking uh, about a company that is trying to begin, you know, say their, their AI journey. Um, is there any advice that you would give to these organizations like where to begin, where to you know, start their research, um, maybe where to put their initial investment dollars in uh, as they go down this path to developing uh, you know, AI, uh capabilities within their organization.

Speaker C: Yeah, well I think that's a very good question and I think there's three steps to, to follow. The first one is education. As I already alluded to, you know, education is crucial. Um, and yeah, uh, pitching futurewise, the company I'm developing, that's an excellent platform to get your education, to get your, you know, your personalized insights on fast changing topics. Um, and, but uh, of course there are many other platforms where you can find the content but it is important not only to rely on the mainstream media because the real insights, the real nuggets, uh, of where the signals, of where the world is heading is happening on the edges of the Internet. And it's not happening in mainstream media. When it's in mainstream media as an organization, you're already too late. Um, so you need to know what's happening from the edges of the Internet. So educate your, uh, organization. And that is not only applicable to your leadership team, very much so, but not only. In fact, it also applies to all your employees, um, because everyone can contribute their own perspective, their own unique way, um, in understanding what's happening. So I think that is really, really important. Once you have started to educate, um, um, your organization, um, then it starts to experiment, start playing with these technologies, start using these new technologies. As I said, there are dozens of new companies, tools, technologies being launched on a daily basis. So that doesn't mean you need to spend hours and hours or millions of dollars on each tool, but it does mean empowering your staff to play with these tools, uh, to experiment with them, to understand how they have an impact on their specific line of work, uh, their specific department, their specific industry. Um, and once you've done the experimentation, then it moves into the execution phase, where you start developing pilot programs, where you start, um, really implementing these tools within the different processes within your organization. And you learn from that, you share those learnings with the wider organization and then you build on that. Um, the cycle continues with more education, more experimentation, and more execution. Uh, I think that's the best way to start, uh, with AI, um, and to be ready for the massive disruption that's coming.

Speaker B: Excellent. Well, on behalf of John and I, we really thank you for being on Unrivited. We will, um, end our recording here in a moment, but, uh, just wanted to say thank you for joining the podcast.

Speaker C: Well, thanks for having me. It's been an absolute pleasure.

Speaker B: Thank you.

Speaker A: Smart.