Don’t Surrender Your Thinking

Speaker A: Wonder about the future and how you'll

Speaker B: be working and learning. Welcome to you in 2042, the future of Work, with your host, Danielle Wallace.

Speaker A: Hello and welcome to you in 2042, the future of Work. Joining me today is John Bennett.

Speaker B: So I've been an entrepreneur now for 37 years. And it, uh, started when I was at school. I used to buy cans of Coke in sort of bulk and I'd put them in my bag and I'd walk the two miles to school and I'd sell them to fellow students. And that's kind of how my entrepreneurial career started, really. And then I left school and went straight from there. And obviously, being an entrepreneur for that long, I've seen pretty much a lot of technological change. I've seen the rise of ecom, social media, and now AI, which is the biggest change of all. So I now spend my time working in AI and trying to help people understand it because there's a lot of noise about AI and it's, I think it's causing quite a lot of confusion.

Speaker A: There is so much change and there is that confusion anxiety at play. So, John, you know, by 2042, AI is going to be embedded in everyday work. Uh, it's going to be here. It's already here. How do you think this will change how people are going to think, decide and exercise judgment on the job?

Speaker B: Well, that's a big question. I think, number one, things are going to change. That is sort of number one. M. I think I see people taking sort of a number of broad approaches to AI, and one is just hoping it will go away. And I think it's not going away. You know, it's. It's here to stay. And I think that means that how it's going to change our work really depends on how we react. I think there's kind of two ways we can kind of go. One is we just mindlessly accept what AI, uh, gives us. And I think if we do that, then we run a real risk of losing our skills, losing our ability to think, to decide, to apply judgment. And I think that's where we really face the risk of becoming replaceable. Then you've got the other approach, which is to really emphasize and develop those skills. So I think if we have a process, if we really think critically about AI, then over time, a, we'll develop our own skills, but B, we'll be able to use AI in good ways and ways that will save us time and if you like, will give us more time to then think critically and exercise our judgment. So it really is down to us. I think, you know, what the future

Speaker A: holds and that future, I mean, that's really what the promise is. The latter is the promise that we keep hearing and seeing, but yet we're still mirrored in the fact that that's not necessarily guaranteed. 2042, we might not be in that route, especially as workplaces get faster and more automated. And it is so easy to lose that thinking skills. I even think of myself, it's now because various generative AI, just so integral and part of my life, I find myself even like surrendering my thinking. So do you think with all that at, uh, play, do you think skepticism and critical thinking then will actually become less common or more valuable or rare in the future?

Speaker B: I think the answer is both. Uh, so I think they're going to be increasingly valuable. I think that ability to think critically, to be skeptical, I call it productive skepticism. So it's that ability to be open to things but also evaluate them critically, I think is going to be increasingly important, especially as AI, uh, becomes more powerful and more prevalent. But I think if we don't exercise them and we just mindlessly use AI, they'll become less common. So I think we're looking at skills there that will become increasingly important, but fewer people will have them.

Speaker A: Do you think that we're facing an atrophy of our critical thinking skills?

Speaker B: Absolutely, absolutely. So the example that I use quite often, I mean, years ago we used to use maps to navigate. So we'd be going on a journey, we'd look at the map, and of course we couldn't read the map while we were driving. So then we put together in our mind, okay, well I need to take this road and then at that junction, I need to take that road. I don't know how many of us can do that anymore because now we just have sat nav or navigation app. For me, my watch gives me, I think it's three beeps when it's a left and five beeps for a right. All that's gone, you know, that ability to read a map. And I think we have that same risk with our critical thinking and our ability to make decisions.

Speaker A: Ooh, that's so interesting. So in this future then, when technology is shaping so much of our work, maybe our over reliance stemming from today now on AI, how do you think people will end up feeling more grounded, human and or even having to acquire or use or maybe I'm not. Those critical thinking skills we just mentioned,

Speaker B: there's a number of elements there. So One is just using those skills. So this is kind of where I try and help people is to approach AI from that perspective of making sure you start with human intent, you really have a good strong understanding of what you're trying to achieve and applying your own expertise and skills to whatever the AI comes up with. So treating whatever AI gives you as a draft, it's not the answer. Using your expertise then to mold and go back and in the end applying judgment, taking responsibility. Am I prepared to stake my name and my career on this piece of work? So I think the first part is that is applying those human elements. But I think the other part that you mentioned in your question was about staying grounded and feeling human. And I think there's a lot to be said for human interaction. So a lot of the work that I do is on how we interact with AI. But I think also at the same time we need to really emphasize our human interaction. So whilst we're working on our ah, better skills at AI, making sure that we're still talking to people, if we're unsure about it, talking to colleagues or talking to friends about it, joining communities and if AI saves us some time, then maybe we just use a little bit of that time to re engage in face to face conversation.

Speaker A: Oh, these are good practical pieces. Firstly, this idea of exercising our critical thinking now when using uh, various generative AI by thinking of as a draft like you said, firstly, secondly, using our judgment, our thinking skills on that entire piece. And then thirdly with time savings is seeking out other human interaction to get validation, using that potential time savings not just to do more of more, like more of more, but actually to create that humanness, the missing piece that is otherwise void within that, that work cycle. Those are very practical ways of approaching this. So with this then it seems like this trajectory of critical thinking becomes not certain. John, like everything you're saying I'm seeing isn't a sure thing. Yes, everybody's going to be exercising critical thinking as they are using large language models right now and everybody's using judgment. No, I don't know if we are seeing people taking this approach right now. What are your thoughts?

Speaker B: I think like everything else in life, it's a choice. We can take the easy way or we could, or we could put the effort in. And I think it's really important to see AI in that context as an amplifier. So if you're thinking, if you're applying critical thinking, if you're learning AI can amplify that. If what you're trying to do is Put the least effort in and get to the quick and easy results. AI can amplify that as well. So I think there really is a choice there that we take the one path or we take the other path.

Speaker A: Yes. So can you give it. I'll share what I think is a practical example. Love you. To validate that, I'll give you your own example. So in terms of that easy path, it might be creating, uh, presentation using a, uh, large language model, just even the text. Large language model to formulate the text, formulate the speaker's notes. It's not quite right. So we ask it. It's not quite right. I don't like this. There's no critical thinking there. You know, it will generate something else. And we do that for a bit. Oh, Then we start to realize, okay, maybe it's the tone is not right. That's not that much critical thinking. But eventually gets to a stage where it's fine, whatever it's been used. That's one path. The second path is generous first draft. And critical thinking is like, well, okay, I can discern from this that, uh, actually the objectives are wrong. Like, the objectives are wrong. This is not the angle. I need to have a tighter reframing of the audience. I actually need to think about who the audience is. Now that I know who the audience is, I can either adapt it myself or I can now use a large language model to refine the audience and their benefit. Focused statement, uh, that resonates with them. Like, is the first an example of relying, of surrendering my thinking to AI, and the second one an example of me using my judgment to go on path two.

Speaker B: I think it's really interesting because there's always an intensity or a flexibility in how much thinking you give. And, uh, actually, I mean, even your first example, there was some critical thinking there. You didn't just mindlessly accept what the AI gave you. So for me, kind of the archetypal examples of mindlessly accepting, or this thing where you see all this stuff on LinkedIn at the moment where there's this whole backlash against em m dashes. Now, I'm British, so we don't generally use EM dashes at all, ever. And I do know that they were used before AI, so they're not an AI invention, as people try and say they are. I've got plenty of books on my bookshelf where there are EM dashes, but they've been used in those old texts with thought. I think what's happening here, why there's this whole backlash is because people are using them in places where they weren't used before because they're just taking the output of the AI, they're not applying critical thinking. So for me, that whole thing where you read the stories of, you know, this thing about the EM dashes or there's just something in the news today in the UK where a police force used AI in a report and it completely made up some results about something that happened at a football match, that to me, that's the mindlessly taking it, you're taking it output from the AI and you're not even thinking about whether it's correct or not. The far end of the example is like the one that you said, where you're getting really, really clear about your human intent, using your expertise in what I call human intervention to go into a dialogue with the results, and then you're using judgment and accountability at the end. So that's kind of like, if you like, the gold standard. But I think with any of these things there's a flexibility and there are times when you can use sort of less intervention. But I think there has to be a decision and a choice. So in other words, if you look at something and you go, do you know what this is? Something I can just put minimal sort of critical thinking into the start, or I can only do a small review because that's all that's needed. Even that's fine because you still use your critical thinking to make that decision. So I think that's kind of what I'm selling, if you like. We have to make a decision every time we interact with AI. What sort of level intensity of me m and my kind of thinking and understanding do I need to bring each time?

Speaker A: That's a good point. And that was my this morning's example, my real life example. So your expertise is very, very helpful on that. Is there anything else that, uh, people can do? So they are not just besides making that conscious choice, anything else people can consciously do. So they are not just blindly accepting what's AI and surrendering their thinking.

Speaker B: I think one of the key things is understanding what AI is. So even the name, we call it artificial intelligence. But most people don't understand what AI. Particularly when I talk about AI, I'm talking about the generative AI models that we use, like ChatGPT and Claude. They're pattern prediction machines, which means they are trained on vast data to understand if this word comes here, then there's a chance that this word might come next. So, for example, they don't understand going for A picnic and getting soaked, or the fact that crops need water to grow. But they know that if the word rain is in a piece of text, they may, maybe the next word might, you know, a word very close will be umbrella or forecast. So they don't understand things. They're not intelligent in the way that humans are intelligent. They don't comprehend. And I think grounding ourselves in that and understanding what they are is super important. And don't get me wrong, intellectually I understand that. But when I'm interacting with AI, I still forget that it's a pattern prediction machine and it's not intelligent. And I'll be engaged in a conversation and I'll feel like I'm talking to somebody, even though I understand logically what it is. So I think just reminding ourselves what they are. I'm sure over time models will change, but right now they're pattern prediction machines. They're not intelligent in the way that we are. I think that helps a lot.

Speaker A: That makes so much sense, keeping the grounding and pattern prediction machine. And then the last question I'll ask you, which is a burning question from every single guest on here, because nobody has quite the answer yet. You have a unique mindset into this. I'm particularly curious, John, what can people be doing to build, to build these critical thinking skills that are needed so we can do the option two that we have shared, but without having critical thinking skills or newer generations, how can people actually build those skills?

Speaker B: The good news is just by using them, I mean, even starting small, I think we need to be curious and be always wondering what things are, wondering how things work. We need to have a growth mindset so not be close to new things, not be close to new ideas, to approach every new thing as a learning opportunity. I think if we do that, if we go into things eyes wide open, trying to find out what things are trying to think about, maybe what assumptions are built into things, or what things are missed, or what context isn't there. I think even if we just do that gently, just small steps at a time, then, um, we can build those skills.

Speaker A: I love that. So we can be cognizant of the fact that large language models, generative AI are pattern predictions to help ground ourselves. We can have this curiosity, this growth mindset, as we are engaging with various gender of AI. So we realize when we're creating, at least at this point in time is a draft. And even in the future, it is always something that requires our judgment to discern if the output is useful and then adding in the element of engaging with other humans, whether it's on that particular complex piece or it's in the world in general. I think all that is what I'm hearing you say allows us to be able to march forward with AI as that partner with us, but not with us just blindly following it into the unknown.

Speaker B: Absolutely. I think you've recapped that wonderfully, John.

Speaker A: Where can people learn more about you and also purchase your book? Don't surrender your thinking.

Speaker B: So the book is available on, um, all online booksellers. But in terms of the other things that I do, the best place really is to visit my website, which is formidably.com amazing.

Speaker A: We'll put that into the show. Notes. Thank you so much for your time and insights. Really appreciate this now.

Speaker B: Thanks for having me.

Speaker A: Thank you for being a part of the future. Subscribe now to stay current.