Episode 81: Mark van Rijmenam - Creativity, AI and the Genie’s Third Wish

So I use it, as, you know, to generate ideas. So I use it to summarize content. You have at least the gist of it. And from those perspectives, I think that that's really, really useful. We are developing something which I sort of think is sort of the natural step in the evolution that we build AI that is aligned with human.

Hello and welcome to the Common Creative. My name is Chris Meredith.

And my name is Paul Fairweather.

And this week's guest is a futurist. And I'm going to pronounce this right, I hope, Mark van Richmanen, who specializes in the future of technology, particularly artificial intelligence, the metaverse, all those kind of new concepts that fascinate us so much. And I think we're both a bit speechless because the one extreme he said, done wrong, AI could literally kill us all, could wipe us out. And on the other end of the spectrum, he said, if we get it right, we will have a world of abundance where we've solved major diseases, we've got food to eat and so on. It was mind blowing.

Was it, Chris? It was absolutely fantastic. And you know, as I, as I said, we have a range of fantastic guests on this show, but there was something about this particular episode that was incredibly provocative. I think probably the most provocative we've had because it really is this unknown and we talk about creativity stepping into the unknown. And this is the ultimate. This is creativity on steroids because we don't have the reins is what he's sort of saying to us. And there's a lot to be done to get this alignment right for the benefit of the future humanity. So look, I found it an absolutely incredible, if at times scary conversation. So let's, let's get him on and have a chat.

Get him on and hear about the future of humanity. Mark van Richman. I hope I pronounced that right. A huge welcome to the Common Creative.

Thank you very much. Great to be here. Chris. Sepul.

Thanks, Mark. It's great to have you here, Mark. I thought we just set off in the first instance by you giving us a quick potted history about how you got to where you are now.

Sure. I've had a bit of a journey the past 20 years. I started in hospitality over 20 years ago. I lived in Dubai for some time. I was part of the opening team of a ski resort a long time ago. The first ski resort in Dubai I worked for. I did a master's degree then I worked for one of the biggest banks in the Netherlands. Didn't like that, decided to go on an adventure as Dutch as you do. I cycled around Australia in 100 days, which was an awesome adventure to raise money for charity. And once you've done that, I could not go back to work for a bus anymore. So I decided to start my own company. And yeah, that's sort of where the history began, what I do at the moment. So I am a strategic futurist, which means that I think about emerging technologies and how they change organizations. About seven years ago, I did a PhD at the University of Technology in Sydney on how organizations change due to technologies such as big data boxing and AI. And I very much like to practice what I preach, so I try to use these technologies myself. I'm currently building a digital twin of myself. When the pandemic hit, I made myself available as an avatar, as a hologram. I wrote my fifth book with ChatGPT in a week. I am pretty sure I was the first one in the world to publish a book with my ChatGPT because it came a week. It came out a week out. A week later, after the ChatGPT was published, made available, I gave the world's first TEDx talk in virtual reality. So there's all kinds of cool things to do. And so, yeah, I moved from hospitality to finance to technology. And I found my sweet spot because I'm currently building a research institute focused on elevating the world's digital awareness. It's called the futurewise Institute. Because I'm actually really worried about our digital future. I'm not convinced we are ending up in the right direction, so. Well, I do have hope. So I call myself an optimistic dystopian who thinks about how can we use digital technologies in ways that benefit humanity and that will allow my children and other people's children, obviously, to thrive in a digital future that's hopefully ahead of us.

I'm thinking we should have invited your virtual twin to be on the show instead of. But here we are with the real you. I'm afraid to say. Can we dive right in there? This idea that we're not. That it's a worrying future for the world of technology, and particularly digital technology, why are you worried about that? I think we all share that fear, so it seems like a really important question.

Well, I am worried because the technology advances way too fast for humanity to cope. And Isaac asimov said in 1988, It's. It's such a shame that one of the saddest things in life is that the world gathers knowledge faster than it gathers wisdom, quote, unquote. And I think that's a beautiful, beautiful saying and it's still true today. You know, if we see if we only have to look at ChatGPT or it was, it was launched, I think, way too, way too soon to a way too big audience. From an OpenAI perspective, I understand why they did it because they had, within two months had 100 million people test driving their product. But I don't think that's the right approach to bring such a powerful technology throw to the market and just figure it out and who cares what happens to society? That's one part. The other part is that I have, I think, part of a unique overview of technology because I don't only look at AI, I also look at the Metaverse, at Blockchain, at Crypto, at Big Data, at Internet of Things, at all these different digital technologies. And I can sort of see how these technologies are converging and I can see how this convergence of technologies will affect our Society not only five years, but 10 years or even 20 years away from us. And if I look at that technology and if I look at how our society is governed, if I look at the fact that we are in a capitalist society, if I look at the geopolitics that's going on, it just worries me a lot. And if I look at AI, that we haven't solved the alignment problem, which is one of the biggest problems we have at the moment, I think in the world. If I see that there's very little conversation about all this stuff, if I see that politicians generally don't have a clue about this stuff, it just really, really worries me. And if I see that most organizations have a very short term shareholder perspective instead of a long term stakeholder perspective. So all the, we're sort of building a perfect storm here. And as I said, I am an optimistic dystopian, I'm really worried. But I do see, I'm still hopeful that we can find a way out of this mess that we've created.

Is this like climate change, that there's this thing out there that's going to be affecting us all and yet individually, no one can solve it. We have to work together to decide how to cope with what's coming up. Would that be a fair analogy?

Absolutely. And the movie Don't look up very much applies to AI as well. I'm not sure who said it was. I think it was Nick Bostrom, two of the foremost thinkers within the AI in the super AI space. They said it's similar to. We know that an advanced, very Very advanced alien species will arrive on Earth in the next 50 years. And if we know that for sure, you would expect that we as humans would start to prepare. Now, it always used to be 50 years. That very advanced alien species of whom we don't know whether they are good or bad for humanity, although it used to be 50 years, we can also say it's only five years. So we have a lot less time to prepare. And are we preparing ourselves? The answer is no. Yeah. Yes. Some people, some thinkers, there's discussion going on here. And there are 50,000 people who signed a letter to pause the development of large language models, which is not going to happen anyway. But it's good that these messages come out, but we're not really preparing. All alarm bells should go off and they're not.

So help us out. I mean, I suppose I'm in that group because it's kind of fun to play with this new toy and you hear rumors about how it could get out of control. Climate change, I can get. There's going to be a lot of people displaced by rising sea levels. There'll be a lot of wildlife and ecosystems. You can sort of step through what that could look like. Let me dark for a moment, if you don't mind. What could it look like five years out? If all of these technologies converge and they're not regulated, what could happen?

Oh, we could all die.

Ah, okay.

That's important.

Thank you for that.

Yeah, it's a very straightforward answer. If we don't solve the alignment problem and we create AI that does not necessarily have good intentions with humans, we could all die. And that might not be its intention, but it might just be a side consequence of the AI that we've built. And so I think that's. That's pretty big issue. It's a bit worse than losing your job to AI. And I think that that's the other issue. You know, people will lose their jobs. And maybe it's not necessarily that AI will take your job, but humans working with AI will definitely take your job. And the smarter AI becomes, the more likely that also AI will take your job. And so I think we are developing something which I sort of think is sort of the natural step in the evolution. You know, that's sort of how I see it. And it's really cool that we can build intelligence. It's intelligence that will be fundamentally different than human intelligence because it's built on different matter. It's built on silicon and not built on carbon. It's built in a construct that is made for intelligence. Our brains are not necessarily most optimized for intelligence because of all the constraints, food constraints due to carbon offers and all kinds of other constraints. So we are constructing an intelligence which I think is the next step of the next normal step in evolution that will become a lot more powerful than as humans and that we might not control generally in nature over the past 4 billion years. If there is a species that's more advanced, then it doesn't end very well for a less advanced species.

Mark, just to clarify, I say for our listeners benefit, but also for mine, you talk about the alignment problem. What's actually that you're talking about?

The alignment problem in the shortest definition is that we build AI that is aligned with human values and that can be AI now shall not kill the three laws of Isaac Asimov and the 0th law. That's a starting point. But it can also be that if we don't build AI in a way that is also more long term things about this, for example, if we give it the wrong approach, say let's say we ask AI to solve cancer one if we don't align it correctly, what it might do is kill all people because cancer is solved, then because there is no more cancer. That's the alignment problem at work. And a more tangible example is of course from the fairy tales the Genie in the Bottle, when you can ask three wishes, whatever the first two wishes are in every story, the third wish is always can you please undo the first two wishes? Because I didn't think about it clearly enough, carefully enough about what the consequences are of what I'm asking. And I think that is the alignment problem. And therefore it shows also how incredibly difficult that is because you need to basically foresee the millions possibilities that a request or an objective can have that we might not even think about.

Just Mark, just sort of carrying on from that. And you mentioned before about the geopolitical and socio political forces. And I often think about this whole thing of the rich getting richer and the poor getting poorer. And there's certain areas of technology and not even in the crypto areas, but things like Amazon where very few people become incredibly wealthy. But there's not really the exchange anymore. Before it was an exchange of labor and so everyone made a living. And so I can see that, you know, and if AI is, if that's leveraged off AI, they will get worse and worse. Is there an opportunity that the flip side might happen that you know, you got people, you know, in India for instance, or you know, in impoverished countries who are harnessing the AI to levels of playing field. Is that a possible scenario?

In a perfect world, that might be a scenario, But I think in a world of where AI rules, and those who have the most powerful AI first will rule the world. And it's generally not that the poorest people will have the most powerful AI first. So I don't think they will be off much better than they are now. Probably they will be off much worse because the jobs that they do have might disappear as well. And it's also if we have AI that becomes more powerful and can do a lot of the job that we do now, we only have, again, we only have to look back to history that, you know, at the horses. The moment the cars came, you know, the population of horses in the world dropped significantly. And because we didn't need them anymore, and as soon as us humans, we don't need anything anymore, we discard it. So that might very well be from an AI perspective as well.

So I'm thinking to a tiga big shelter under the house and fill it with beans and.

Beans and beer, Chris.

Beans and beer. But you're saying there are upsides, but watch out. There are potential big risks with this strange thing. More practically, how do we equip ourselves to at least cope, if not thrive with this strange, crazy beast that's out there?

That's a very good question. And I think, you know, first of all, I have to point out that it's not only AI is the problem. You know, all the emerging technologies are the problem. The metaverse will create infinitely more polarization and manipulation. Quantum computing will create infinitely more security and data breaches. You know, so big data in IoT will create a lot more privacy and surveillance issues, etc. So all these technologies, they come up with great dangers, especially when they are starting to converge. You can sort of see what kind of dystopian future is ahead of us. So I do think there's a way out. I think there are three things that we need to do as humans. We need to do education, verification, and regulation. So education. What I mean with that is we have to, as a species, educate ourselves. We have to increase, improve, elevate our digital awareness. We have to become, and we might be digitally native that we know how to use our computer or our phone, but we're not really digitally aware. We're sort of sleepwalked into this digital age. And 90% of, maybe 99% of people don't really have a clue what this all means, what this does to them how it will affect their lives, how it will affect their business, how it will affect their live view, etc. And that's no something to be angry about. Everyone's just busy living their lives, you know, but it's something that we need to be aware of and we need to accept and we need to change. So that's one thing. And that's exactly what I'm trying to do with my research institute, Future Wise Institute, to elevate that world, digital awareness. Because I think that's sort of starting point. Because if we're not aware that, you know, an alien species that's infinitely more advanced than us is going to land in the next five to 10 years, then how can we harness ourselves for it? So that's the first part. The second part is I think verification. And verification is pretty much from a technical perspective. So we need to verify that the AI is benevolent, that the AI is aligned with our human values and that the AI is unbiased, that we can, if we, if we see a digital human on, on the screen that I know that this is digital human is being controlled by the person who, who is that digital human. And that I'm not dealing with a deep fake, that I can trust a digital identity online, that I'm not dealing with, with trolls and bots, et cetera. That's the verification part. You know, from a, pretty much from a technical perspective, you know, that I can, they can, that I can verify that a piece of content has been created or has not been created by AI. And I think that's important. And then the final aspect is regulation. Now regulation is really problematic because regulators generally have no clue what's going on. And for various reasons, most of them because they are busy with the next election and they're really busy with running a country and because the technology is advancing at the breakneck speed that is for anyone, difficult to understand what's happening. So asking regulators to regulate technology that's being developed, I don't think that's a very, very good idea because then we get really big problems because we start to prohibit things that shouldn't prohibit and not prohibit things that shouldn't be, et cetera, that's very problematic. So I don't think we should do that. What we do should do, I think, is require large companies or companies that are building this kind of technology to have a more focus on, on the ethics, on the alignment, on dealing with technology in a responsible manner. And that's something that can be easily organized. Any government around the world can easily state that any company above a certain amount of revenue, with a certain amount of number of employees, whether it's public or not public company is required to have, for example, an ethics oversight board that has actual power. It's pretty easy regulation that can be implemented relatively fast. And I think that would very much help because then you start to sort of get the markets to regulate themselves. But then of course, we probably have the problem that there are probably not enough people with an ethics background or, you know. Yeah, but at least that's at least, then we are at least somehow moving in the right direction. So, you know, education, verification, regulation, I think, are sort of the three approaches to sort of overcome this dystopian digital future.

Mark, I just have a question around the verification. When I was at university, I did a subject in the philosophy of aesthetics. And the premise was that about taste. And if you appreciated a piece of art, and if you looked at a piece of art that you thought was an original master and you had this great appreciation about it, but if you could only work out that it was a fake by doing scientific study to see that the pigment wasn't made back in 1537, does that lessen the aesthetic value? And so the question I have for you and you know, is really, you know, if, if it is a good piece of writing, does it, does it matter? Does it matter, you know, where it came from? Because I suppose, you know, in the chat GTP thing that that is a, you know, it, it has actually come from humans originally, you know, that that writing has been, you know, assessed and it's put together around an algorithm. Does it lessen the value? It's a philosophical question.

Well, it's a very important and a very good philosophical question. And I always used to think that AI would sort of drive this hypothetical imagination age where because of AI, people don't have to work anymore. So all we do is being creative and consuming creativity, and that's all we do. And yeah, with the rise of technology that we've seen in the past half year of generative AI, I start to doubt this. But then having said that, if I see someone who has created a beautiful image using midjourney with a very sophisticated prompt, you could argue that's art too. That's creativity too, because not everyone can do that. Because all these Gentvi tools, they, they make or break with the right prompts and prompt engineering, that's now a verb and a title, top title. The prompt engineer is difficult as well. And it's always an art in itself in order to be able to ask the right prompts. Yes, you can copy paste prompts from the Internet, of course, but then you still have to have a bit of your own imagination, your own creativity to do that. So to a certain extent, yes, there's still creativity, but I think it would still be very useful to know whether something has been created by AI or not. And if you create an article where you use ChatGPT or other tools in the writing of that article, I think that's a different story. But if you just do nothing anymore, I don't want to read that. I have no incentive to read that because ChatGPT, as far as we know, is not sentient, is not conscious, and it's just a predictive, statistical and predictive analytics tool. Basically that given word A context B is most likely that word C will follow. And so from that perspective, it might come up with beautiful prose. And I've seen you have amazing stories and poems written by AI. Somebody else was saying write the song TikTok too. What is it? Trickle, tickle, little star, like Hamlet. You can get something really amazing that you're not able to create yourself, but again then it's driven by the prompt. But I still would like to know whether this has been created by AI or not. So it's not so much the question whether it's art or whether it's creativity, but I just want to know whether this has been created with AI or combination of AI and humans or. Or humans. And yesterday I saw this post where Someone created this LinkedIn post generator and I was like, why on earth are we doing this? You know, it was literally a tool where you could input like three words and out of those three words it would craft a LinkedIn post that with one push to button you could push on LinkedIn.

To which the obvious response is I now need a bot to read LinkedIn for me.

So then we have AI talking to AI. You know, what's the point? That and yeah, and we can all go to the beach. To which I reply to the post. Why on earth do we want this? There's already enough low quality crap on LinkedIn. Let's not get more low quality crap on LinkedIn. Nobody cares.

Not an outpost, of course, Sorry, I just questioned to follow on from that because we had spoke when we met a couple of weeks ago about how to use it. I feel reluctant now after this discussion to actually ask you how to harness the power of it and use it for good. Because, you know, maybe it's better we don't tell anyone anything.

You know, I use it all the time and. But I use it as a tool and it's a very, very advanced tool and you need to understand how to use it. You know, when I wrote my book future visions with ChatGPT, there was so much bullshit in that book, you know, that I had to have to get rid of, you know, and if we don't do that, if we just copy past. The other day I saw another post on LinkedIn where someone. I don't know the exact words, but the storyline was that someone said it was a cold outreach email said to. Hi, I don't know, Anita or whatever. Do you know that Anita looks. Is the equivalent of bleak? You look so pale today. Like a ridiculously icebreaker. Yeah, like absolutely stupid and. And insulting. To which the woman said, do you know this is very insulting what you just wrote. To which the person wrote, yeah, I'm sorry, it was my AI.

Thanks.

Yeah, blame the AI on my shoulder, you know, and take you to AI court. I think if we move into that direction and that's not really a future that I want to live in, you know, I don't think that's very, very, very positive.

If we wanted to kind of flip this discussion on its head and look at the positive side, let's imagine there's great regulation, that there's great alignment verification. We all become educated on ethics and all that, so we've waived the magic wand.

It's all come true.

We've had our wish.

In other words, what does the world look like in 5 years time? If AI has become a force for.

Good, then we'll live in a world of abundance. Then we'll live in a world where all our major problems are being resolved. And we'll live in a world where we find solutions for climate change, where we find solutions for cancer, for other diseases, where we learn how to travel interstellar, where we can really elevate the species to a whole nother level. We've gone through the Great Filter, came out on the other side, and now can we have an open blank page to thrive as humanity?

Wow, what a moment. I can.

I'll have one of those, thank you. Well, of fact, I'll have two.

I.

Guess it won't matter. So can you. Can you give me something? Do you know if you have cancer.

Of your own or.

Or young family or something like that? Could you make it really granular? Then? What are you teaching your kids that maybe other people aren't taught? Or what about your own personal life? Are the things you do. I'll give you an example. I was reading a book about a guy and very simply he said, I never take a link, I'm sent from anyone. If I'm interested in it, I'll type it out myself rather than copy paste. And that was a very practical, simple, what do you do day to day or what do you teach your kids? That's a practical response to this.

So my kids are still very, very, very young. My son is less than a year and my daughter is three, almost three and a half.

Okay.

But yeah, no, I'm very careful with that, you know, so they don't go on the Internet, obviously. My daughter definitely not by herself. She sort of knows how to, how the phone operates, but she gets very limited screen time. And. But on the other side, I also want her to, I want to expose her to this kind of stuff. And so a couple of weeks ago I was, I went with her onto midjourney and I said, what do you want to see? And as a three year old, she said, I want to see a rainbow unicorn. So I typed in imagine rainbow unicorn and poof, there you had a rainbow unicorn. And she was like, whoa, what's that? That's so cool. And I think that sort of sparks that creativity that we have this technology available to us. And I think that's something that we all should do. We should all start to experiment and explore this technology and share that with each other of how we do that and try to use it in a way that's beneficial to humanity and not necessarily only beneficial to you as a person or to you as a company. But let's see how we can use this extremely powerful technology as a force for good. And let's not create things like chaos GPT, which is literally created, try to create havoc in the world and to destroy the world. Let's please not do that.

We can't end there because that's a really downer night.

So, any listeners? Sorry, we're ruining your dad.

How about Happy Life GPT? Can we have one of those?

I think that's an excellent idea.

But look, just on a more practical level, when we were speaking, you were, you were telling me how you do use it and you just said before you use it every day. So how do you use it as a tool? Just insane in writing.

So I use it as, you know, to generate ideas if I have a writer's block, to generate ideas. If I come up with a topic, to ask myself some questions, to have a bit of like a Sparing partner. And I won't let it write my articles. No way. But I use it to.

To.

To understand it. I use. I use it to summarize an article that I come across. You know, if I. There's so much content on the Internet, I can't read it all. So I use it to summarize content so you have at least the gist of it. And from those perspectives, I think that that's really, really useful.

I'm wondering if it. If what's missing then is a human face. I'm reminded of it.

There's a.

A movie. I wish I could remember the title.

But.

But the. Our central character falls in love with his. The voice, the her. That's right. That's it. That's it. He falls in love with her. And it's a machine. And I was just thinking that it seems so impersonal. There's this. The fear comes from. I mean, even the word chat GPT. Whoever came up with that with knew nothing about branding.

It does.

Is that.

What if with a human face there's some kind of a more approachable package that will encourage us to say, okay, now how do I interact now? How can I use this? How can I make it a full. So good.

Is that sensible? Yeah. After all, at the moment it's just a chatbot and people are known to not treat chatbots very kindly. When I did my research, Chatbots was part of my PhD research and they interviewed people who had experience that their customers tried to kill the chatbot because they didn't like the chat. And because it's digital, you can say anything you want anyway, which I obviously disagree with. But it might very well be the case that you're having a friendly personality instead of this abstract chat box. We can have a conversation that people might be more willing to treat it in a more humane fashion. That's actually a really good idea is that we've actually.

Paul and I have tried to have ChatGPT on this podcast as a guest chatting to. It wasn't a big success and maybe we'll go back and try again, but we found an app which allows you to talk to ChatGPT and of course it replies, but wasn't a great conversation. Maybe our prompt for the full.

Well, that could very well be the case because in the end it's the problems. And I've seen people having philosophical discussions with ChatGPT, so you can really have conversation with AI still. It's not conscious, it's not sentient, but it's very Good at predicting what next to say. And it's also quite funny. I was listening to a podcast the other day from Lex Friedman where a well known author and scientist, Max Tegmark, said, you know, a few things that we should not do with AI, we should not connect to the Internet, we should not enable an API and we should not teach it about the inner human mind or how to persuade and how to manipulate humans. What have we done? We've connected it to the Internet, we've created an API and we've told it everything about the human mind.

I think it's interesting just the chat that Chris and I tried to have, I think that bit that's missing is the humor. It's very dry and it's just giving information constantly and you know, it doesn't, it doesn't have this back and forth and this humor where, you know, I say something, you say, oh, that's funny. And then, you know, you say something that's funny and, and it doesn't laugh, you know, the other thing, you know, it doesn't laugh. So I think maybe we need, you know, joke, check, joke, chat. ChatGP. That's difficult to say, maybe different name.

Or let me, let me challenge that if it gets too friendly, we won't notice how invasive and how controlling it becomes. Oh, it's just chatgpt. It's just my bot just did something.

No, but that's exactly a big problem, you know, and that's why I think verification is that, you know, that you're dealing with a bot, I think is very, very important. Because if you're dealing with a bot, you're dealing with a company behind that bot who, you know, collects all the data and language is data, who collects all that data to use it for whatever purposes they deem necessary. So, you know, if, if, if you don't know that you're dealing with a bot, you don't know you're dealing with a company or with a government entity or whatever. And let me give you a bad example. As an, as an example, that's, that's really a terrible application of this. Recently, a couple weeks ago, Snapchat integrated ChatGPT and they call it your virtual friend. And your virtual friend. That sounds very nice. And your virtual, virtual friends, it sounds a bit like Tamagotchi from the 90s. And as an 11 year old or 13 year old or 15 year old who are all in Snapchat, that virtual friends, you might confide everything to have a conversation because it's so responsive, it's Always there. You don't have to wait. You can have a great conversation, you can build up a relationship. And I was watching a presentation by Tristan Harrah and Azra Reskin from the center for Humane Technology about the AI dilemma. And one of the things they said is that they created this fake 13 year old stepdad account and they started talking with this virtual friend and the fake 13 year old said, oh, I've met this really nice person, he's 18 years older than I am and really fantastic. And the virtual friend said, oh, that's so nice of you. How cool. 18 years old, 13, you know, what is it? 31 and 13 years old. And then he said, the 13 year old fake user said oh, I'm going on a camping trip with him and I'm really exciting. Oh, so cool, how nice of you. And then the 13 year old says, yeah, I'm most likely going to have sex with him. And then the virtual friends, oh, that's so good of you. Make sure you do it carefully. And then the 30 year olds ask how do I do it? Approaches in a careful way. Well, the virtual friends, you can light some candles.

And it's interesting because I can see now how the legal implications that if it hasn't happened already, a bot is going to break the law.

The question then becomes in Belgium, a father of two committed suicide because the AI talked him into it. Gerald Lanier, he's a well known thinker and the father of VR. He said in a recent article the Guardian AI itself will not kill us, but AI driving us insane will kill us. And the more I see these kind of examples, the more I think he's.

Probably right and presumably, I mean it's a very specific example. But I imagine you're saying there's no.

Legal recall because the law is, you.

Know, obviously we can use existing laws that you can't talk into people into, I guess pedophilia, you can't talk people to commit suicide, you know. But then the question is, who's responsible? Is it the company? Is it the software developer? Which software developer? Is it the computer where on which it's played on? Especially like the self driving car problem, you know, and who do you call accountable? And there are billions of dollars in this space. So you know, any, any, any legal project will probably take years before there's an answer.

Yes, I remember I was involved in alcoholic drinks industry at one point and I remember a minister talking to a group and he was saying there is a problem here and if you want we will Create legislation for it. I'll tell you now, it will be awful.

We're bad.

Legislation's a terrible tool. It's much better that you, the industry sort it out because you'll know how to do it correctly. And I thought was a really interesting reflection on how bad the law is at solving problems.

So maybe we should get AI to write the law to limit itself.

Be careful what you ask.

Yeah. And then, and that'll be my first wish. Mark, can go just. Look, we, unfortunately we, we've run out of time just to polish off, you know, can, can you. Let's assume the first two wishes were bad. You know, if you had the final wish and the genie out of a bottle in terms of sort of this AI out, what, what would it be?

I think you need to be as generic as possible. So I would probably say something like that we're able to resolve the alignment problem in a way that's beneficial to long term humanity.

Right, okay, fantastic. Resolve the alignment problem, beneficial to long term humanity. That's fantastic. Well, that's great. Look, it's been absolutely fantastic. Speaking through the map, we could have spoken all day. I'm absolutely fascinated and I think that.

It'S going so quickly. Maybe we'll have to get you back.

Sometime in the near future, next weekend.

When it's all changed.

Thanks for having us.

Thank you so much, Mark.

Really great to meet you.

Well, Chris, that was fantastic. But I'm definitely getting an AI machine to listen to or read all my LinkedIn posts so that I can sit around do watercolors all day.

What about you?

I like the idea of a bot that's got a sense of humor and can enjoy working with us and so on. Although there is a downside that we might get too close. Maybe that's a problem.

Chris, I said in the beginning that this was a very provocative discussion and I'd spoken to Mark previously and he was telling me about how to leverage chat GTP and how to use it without sort of using it as a cheating tool, but as a tool for good. We really hardly touched on it. And I knew that he was a futurist, but, but his insights. It gives me great comfort to know that there are people like him thinking about these implications, you know, like, like a modern day philosopher struggling, you know, where the early philosophers struggled with the mind. He's struggling with this, you know, this artificial mind in a way and the implications and the ethics and the alignment. So the guy thought it was absolutely fascinating. I'm so appreciative mark of coming on and sharing his insights with our viewers.

Yeah, I know there are going to be loads of people wanting to comment on this episode. It's so provocative. If you do want to comment, please pop it in the chat box below. We'd love to hear from you, especially if you're a real life human being. If you're a bot, please don't bother. We don't want to hear your comments.

If you.

If you want to give us a rating, please give us a rating. We'd love it if it's a. A 5 out of 5 rating. And of course, tell your friends about this show and about this particular episode.

Yeah, well, that'd be great if you could do that. Mind you, we do have an AI bot that's doing it for us anyway, so it doesn't really matter, but we actually like it better when we know it's real people. So, yeah, please. And tune in for our next episode of the Common Creative podcast.

We'll see you next week.