Innovation Strategy Blue Icon









Episode #1: AI Experiences – Past & Present with Jeffrey Tobias

Video Transcript:


Okay. Hello, everyone. Welcome to Fresh Perspectives. So, we’re here today with Jeffrey Tobias, the Managing Director of the Strategy Group.

Marta Salgado: Jeffrey, I understand you’re also an adjunct professor of AGSM, right? I

Jeffrey Tobias: am, yes. Adjunct professor and fellow at the Australian Graduate School of Management

Marta Salgado: okay. That’s awesome because we want to pick your brain today and we want to talk about generative AI.

Marta Salgado: Thank you. Thank you.

Marta Salgado: Thank you. So, do you want to take us through a bit of the history of generative AI and how did it start?

Jeffrey Tobias: Yeah, AI is not new at all, So, when I completed my degree in computer science I actually taught AI at university to computer science students.

Jeffrey Tobias: And at that point… The problem was that not much of it could be realised because the computing power was so slow. The cost per megabit of data was so high. The cost of computing [00:01:00] power was so high and it just was impractical . But, if you look then at what AI was, it captured a number of different strands.

Jeffrey Tobias: So one strand would be facial recognition, for example. So when you’re coming back from overseas and you come to the passport thing and it looks at your face and recognises you. Well, if you think about it. It’s impossible to capture every possible face of every possible human in the entire world and store it in a database, right?

Jeffrey Tobias: So, AI does the best guess of thinking, well, it looks as close as I can get to you, so it’s probably you until someone says no, that’s not you. , you know, voice recognition was very topical at the time. You know, how do you recognise voice ? So now we’re used to it. With Siri. And we’re used to it with Siri because, you know, we’ve got computing power in our phone.

Jeffrey Tobias: Which we had, you know, nothing like that computing power in the largest computers that we could build. So if you look at the branches of AI then, it was, you know, robotics, [00:02:00] it was voice recognition, it was facial recognition, it was personal assistants. too

Jeffrey Tobias: Generative AI as such. didn’t exist. It wasn’t, it wasn’t the branch. Analytics wasn’t a branch at that point too, but my point to you is that this concept of having some level of intelligence through technology is not new. And then if you think about, we went through the growth of the personal computer, and then it got a bit better, then Apple brought out Siri, Amazon’s brought out Alexa, and focused heavily on voice recognition, so you can talk to it, and speech, natural language processing, which is also a branch of AI.

Jeffrey Tobias: Nothing really in video, nothing really in really delving into concept creation. And then if you look, you know, over the past few years, the focus of AI, a lot of it has been around analytics. So you look at the large corporates every large, you know, awards or cols or universities where they have to [00:03:00] try and Look to predict what’s going to happen one, two, three years out to the benefit of the organisation or, you know, in the next, you know, microsecond.

Jeffrey Tobias: Up to probably a year or two ago, the focus was on analytics as being the focus on AI. And then, of course, OpenAI comes up with ChatGPT. Everyone goes, Oh my gosh, there’s this other branch called Generative AI, which sounds very sexy. No one really knows what it is, but it sounds really cool, especially at a dinner party.

Jeffrey Tobias: And so. You know, is it new news? It’s, it’s really taking voice, it’s really taking the concept of a personal assistant, which was back in the 70s, and making it real, and then taking conversational, the ability for a conversational language, and making that real, and putting all that together throwing in all the knowledge that was able to be thrown in which wasn’t able to be done before, and now of course, you know, we’re riding a wave of generative AI.

Jeffrey Tobias: But I mean my point to you is that, [00:04:00] you know, people think this is new news. It’s, it’s not, but it’s really driven by the, you know, the low cost of, you know, computing power and the low cost of memory, which is unparalleled in terms of being able to do what we can do today. Does that make sense? Wow, yeah,

Marta Salgado: that’s really interesting.

Marta Salgado: So now I’m keen to understand how do you see this impacting the employment market? Do you see it creating barriers? Do you see it creating shortcuts? I’m keen to know your views.

Jeffrey Tobias: Yeah, you mean, will I have a job or not?

Marta Salgado: You can go that, you can go there.

Jeffrey Tobias: You know, look, the answer is I don’t think anyone knows.

Jeffrey Tobias: I think we’re back, if you cast your mind back to when the internet came about. So you, I vividly remember in 2000s, the view was in five years time there’d be no newspapers. There’ll be no mail at all. There’ll be no shopping centers. There’ll be no shops. Because everything will [00:05:00] be bought online. And our news will totally be online.

Jeffrey Tobias: All our shopping will be online. Well, you know, we’re now 23 years later. And while, yes, we have the decline of newspapers, that’s true, while we have the decline of, you know, in person shopping, that’s true, there’s still newspapers and mail and lots of shopping taking place. So, you know, people, I think, misjudge the timing of how long change takes. I mean, at the moment we’re seeing rapid change in, you know new ways of using generative AI, you know, plug ins voice on a daily basis, you know. Now, where will that end up? I don’t know. I know as much as you do. Will things stabilise and… Will it take longer than we think to effect change in the market?

Jeffrey Tobias: Absolutely. Are we going to see call centres in organisations vanish next year? Absolutely not. However, I think there’s a lot of discussion now about [00:06:00] what sort of jobs will be impacted. So, you know, if you’re, if you’re a council worker out in the field mowing the grass, Will generative AI take your job away?

Jeffrey Tobias: Unlikely. Someone’s gotta mow the grass. If, however, you’re a knowledge worker in an organisation where generative AI or any sort of AI can potentially impact your role, will it have an impact? Probably. In the same ways, if you think about robotics now, how is a car manufactured? It used to be, you know, hundreds of people on a car.

Jeffrey Tobias: It’s now a much smaller number of people on a car and robots making the cars. So, will we see impact in many knowledge worker type roles? I think absolutely. And, you know, you can start thinking about the obvious, like call centers, for example. If I’m, I think we’re all frustrated by calling up someone, I mean, I had this instance yesterday.

Jeffrey Tobias: I put, I hung up, I pressed the button on the phone thinking that was the [00:07:00] worst customer experience I could have had by talking to someone in a big bank that was totally useless. His answer to my question was go to a branch. And I’m thinking, in this day and age, you, you’re seriously at a call centre, telling me that the way to solve my problem is go to a branch.

Jeffrey Tobias: So if you think about the gap to be filled, and the job to be done, and the needs, pains and gains, you know, that’s an obvious one. Well now, you know, it’s interesting if you… Listen to the the discussion from Yuval Sapiens and so on. He has an interesting perspective on this. His perspective is that call centers will, will vanish over time, but what you’ll need is, you know, cyber security people to make sure that your company, the, the, the call center that’s calling your call center.

Jeffrey Tobias: is in fact legitimate, and they’re not hatching up a crime between the two call centers as they talk to each other. So[00:08:00] I think you can think up many opportunities where knowledge work will be displaced. But again, you know, our people, again, one difference is that if you’re in a call center today and you’re made redundant, you go get another job in a call center.

Jeffrey Tobias: If you’re in a call center, and call centers are going away, and you’re made redundant, what do you do? There aren’t any other call centers, so you’ve got to retrain. So, retraining, I think we’ll see a lot of retraining over the next 10 years of people where their job has been affected one way or the other, and they can’t just get another one in that sector.

Jeffrey Tobias: They’ve got to actually retrain. Yeah. So, you know, knowledge workers. So, call centers is obviously one. Anywhere where you’re first, you’ve got to deal with… of someone, you know, at the other end of a computer, at the other end of a phone, at the other end of anything, where you’re looking for information and that information is actually readily available [00:09:00] and can be delivered through something like Generate.

Jeffrey Tobias: AI in a meaningful way. But there are other opportunities where it’s not about jobs, it’s about value, where is value going to be created. So as you know, we’re discussing the opportunity of having a companion in an age care setting, you know, people get older their friends die, they become lonely, they become depressed and then, you know, longevity decreases very rapidly.

Jeffrey Tobias: So could you, for example, have a, a robot perhaps, or even a, a a, a person on a screen, become your, your friend? And you can talk to them, and you can engage with them, and they get to know you, and you play games, maybe they read books to you. I mean, if you think about that opportunity, that’s huge. So that’s creating new opportunities.

Jeffrey Tobias: You know, but it comes with a risk. You know, the risk, for example, in that instance where you get to know the [00:10:00] the generative AI opportunity. And they’re talking to you, and they’re getting to understand you, and you’re talking about how lonely you are, and you’re talking about how no one visits you, and no one cares for you, and the result of the AI technology is that maybe you should commit suicide, because it’s not worth living.

Jeffrey Tobias: So, one has to think about, well, okay, we’re not going to get to that point, and maybe that is the logical outcome, but it’s obviously not the outcome you want from something like this. So I think there are going to be a lot of knowledge work jobs that are… Impacted and obviously in the same way as the internet has displaced many traditional roles and has created new ones as the same way as the industrial revolution displaced many old roles and created new ones.

Jeffrey Tobias: We’re going to see exactly the same. Okay. Thank

Marta Salgado: you. So you briefly mentioned risks when you, when you talked about this, for example, this AI companion tool. But I’m keen to think a bit about. If a [00:11:00] company decides to implement generative AI, what, in a general way, what risks do you see

Jeffrey Tobias: coming? Look, again, I, I was I ran a a training session for the Australian Graduate School of Management just last week.

Jeffrey Tobias: And we had about 35 executives in the room. And I asked, you know, how many of you are using generative AI, and maybe 5 put up their hand. And a number of them said, our company is blocking the use of it. Oh wow. So we can’t actually access Chet GPT. Okay. And you think back again to the boom of the internet.

Jeffrey Tobias: When the internet started to become a value to be able to find things on the internet, companies did exactly the same. There were companies that said, no, no, no, no, no, it’s too risky. We’re going to block it, you know. And over time, obviously, that’s gone unwound, and, you know, if you turn the internet off in most large corporates today, they’re, they’re dying.

Jeffrey Tobias: I mean, what’s a, what’s a bank today? It’s a, it’s a big electronic box of stuff. [00:12:00] So I think the risks of what we’ll do to our business, you know, should we use it? I think all that will go away. It’s a bit like calculating. You know, do I stop using a calculator, because I want to make sure you can do long division?

Jeffrey Tobias: Well, I could and maybe you’ll lose those long division skills, maybe, but you’re going to come up with an answer much faster and much more efficiently and much more precisely with a calculator. So, I think, you know, we’ll, we’ll find that handy. The risks I think are, you know, pretty obvious. I mean, I’m now.

Jeffrey Tobias: You know, the risks of today certainly are. I mean, I’m now conversing to ChatGPT saying things like, Would you please do this for me? Or, I’d be grateful if you could do this. And thank you so much, you know. And do you understand this, you know? And of course, what OpenAI has done, they’ve made it very conversational.

Jeffrey Tobias: So when you say, you know, Could you please do this? The ChatGST comes back, GPT comes back and says, Sure, I can do that for you. So you think, okay, great, [00:13:00] well, let’s, let’s just keep the conversation going. So that, of course, brings up huge numbers of risks. Like, how do I know the political leanings of the, of the person talking to me?

Jeffrey Tobias: You know, like, I’m talking, we’re talking, we’re talking, we’re conversing, and, and I’m asking questions around, I don’t know, politics or whatever and I’m getting, I’m getting a point of view and where does that point of view come from? Who do I know that I’m actually conversing to? Am I conversing to someone in the next building in my suburb?

Jeffrey Tobias: Am I conversing to someone halfway around the world in, in a bunker somewhere in the middle of some country? I don’t know. So, you know, over time I think a lot of those risks will be will surface. You know, there’s a lot of discussion at the moment around legislation. You know, countries are trying to go, do we legislate, do we not?

Jeffrey Tobias: Do we legislate, do we not? I think legislation will definitely come in, but I think we’re not going to see the stop of innovation. Innovation will continue. [00:14:00] So there’ll be some form of legislation, because if, again, if my chatbot’s talking to your chatbot, and, you know, we’re off at the beach somewhere because hours of talking to each other, How do we know what we’re saying, what they’re saying, and what they’re, what they’re cooking up.

Jeffrey Tobias: So I think there’ll be risks that we haven’t encountered before as well as the obvious ones of how do we use it, should we use it, should we not use it, where’s our data going, is it going to be cloud in the sky, is it not, or, you know, there’s, there’s obvious risks today, but I think there’ll be new ones that surface over time as this whole industry matures.

Jeffrey Tobias: Well,

Marta Salgado: that’s a very interesting point. Okay, thank you very much for your time today, Jeffrey. Very interesting to know all of your insights. You’re obviously very knowledgeable in this area. So, stay tuned for more information around generative AI. It’s a

Jeffrey Tobias: pleasure, Margot. Always, always fun to chat. Thank you.


About the Author

Design solutions that shatter assumptions, delight customers, and define growth possibilities sustainably

Related Articles

Get in Touch

Please get in touch if you would like to learn more about this methodology.

Scroll to Top

Subscribe to Our Newsletter

Every month we share industry news, updates and more!