Episode Transcript
[00:00:00] Speaker A: Welcome, welcome. This is Rob. And today we have a special bonus episode of the Business English Podcast featuring a crossover with Simple English News Daily, or send 7 for short.
Today, Stephen and I discuss artificial intelligence, its current impact on the business world, its potential future implications, its applications in English language learning, and much more.
What you're about to hear is the second part of our conversation. To hear the first part, you can find it released today on the Send7 podcast feed. A link is available in the Show Notes for your convenience.
Now, the Business English Podcast is of course free to listen to, but for access to premium content, including full transcripts, vocabulary lists, worksheets and exclusive episodes, consider joining the Business English Podcast premium subscription. The link is also in the Show Notes.
Now let's dive in to my conversation with Stephen.
[00:01:11] Speaker B: Any initial thoughts on this? Any opening arguments?
[00:01:14] Speaker C: It's one thing which I've heard other people think about, I should say about AI taking jobs is that it may be strangely the, how do we say, the more graduate level, higher, typically higher educated jobs that might actually go first. And I've seen this example given of the hospital in which a doctor who is analyzing test results, that is a job which could very probably be taken by AI relatively quickly, whereas the nurse who is physically giving an injection and cleaning somebody, turning somebody over, making a space usable for patients and running around the hospital very quickly, that is something which is far more difficult and less likely to be taken by AI anytime soon. Yeah, that was something that I've thought about.
[00:02:05] Speaker B: Yeah. This idea of everybody thought it may take the lower grade workers out, potentially blue collar or some sort of lower grade knowledge workers, when actually to your point, it's actually challenging the professional white collar areas far more aggressively.
We know already that multiple companies are integrated it into middle management and to back up at least the decisions that are being made by management. What I think is rather interesting and I don't know if you're a Star wars fan, Stephen.
[00:02:37] Speaker C: I'm not.
[00:02:38] Speaker B: Oh, Are you aware of a small green thing called Yoda?
[00:02:42] Speaker C: I am, I am.
[00:02:43] Speaker B: So this idea of the Yoda, the sort of supporting element to Skywalker, the co pilot, I feel that this is the best way to imagine AI in its current form is that whoever you are or whatever you do, you're still the star of the show, you're still the core component. And I think without having any copyright strikes from Microsoft, they, I think they've named theirs Copilot or something like that. But that's exactly how I envisage AI at the moment, is that it's an enabler.
[00:03:12] Speaker C: Okay, does that mean that you don't think that people are going to be losing their jobs to AI, then you just think they're going to be assisted, but they're not going to be losing their jobs?
[00:03:22] Speaker B: Not in its current form. I think we're still looking at it as a co pilot. There are, I think there's a multitude of things that need to happen for that gap to evaporate completely. We looked at it in the Industrial revolution. Yes, okay, people stopped plowing fields by themselves, but fields still get plowed and there are still people taking care of the tractors and driving the tractors and making tractors, the motor car, etc. Etc. So I think there will be new waves of jobs created by the displacement of people or having their jobs optimized or streamlined by AI.
So if your job does look like it may be taken in the next five or ten years, this is, it's a great warning sign now to say, okay, maybe I can do something and get on board with this AI because it's not going anywhere. People are trying to ignore it, but it's not going anywhere.
[00:04:12] Speaker C: Yeah, I'm happy to say that it doesn't seem to be able to do my job yet. I've actually tried. As most people listening to this know, I make a seven minute podcast every day of world news. And I tried having a go and seeing what happens if I go on to something like ChatGPT and just say, give me like a seven minute script of important things that have happened in the world today. And I've tried changing the words a bit and it's just not very good. I don't know why, but the stories it chooses are just not stories. I would choose the way that it phrases things. It just doesn't sound right. If I wanted to use it, I would have to change every single sentence. It also, it's not new enough because the things that I'm talking about are things that have happened in the last 24 hours and the things that it gives me just are not that new. Even if I say make sure that it's coming in the last 24 hours, it's just not, I don't know, it just, it's just not quite good enough. And that is even before thing that's just thinking about the script, that's not thinking about the.
Obviously turning it into a podcast and distributing it all over the Internet.
Yeah, I think luckily for this podcast, I think I've got a few more years left of life in my job.
[00:05:21] Speaker B: This is Fantastic news, I might add, Rob.
[00:05:25] Speaker C: That is despite the fact that I've received a few emails in the past accusing the podcast of being made by AI. I've actually even had one, I think it was a teacher write to me and say, I've put your podcast on in my class. And some of my students have said, oh, that's made using AI. I just wanted to confirm whether it is or not. And I had to write back saying, no, it's not. We don't say anything. I know.
[00:05:48] Speaker B: Disheartening hours and hours into it and somebody says, oh, no, that's fake. It's not really.
[00:05:53] Speaker C: Yeah, it's just made by a robot. Nope.
[00:05:55] Speaker B: Yeah, it's a pretty challenging situation. But as they say, when you're a hammer, everything's a nail if you've only got this lens to look through.
So actually, no, I spend hours and hours scripting and researching and implementing. I think you can normally tell. Definitely at this stage, you can normally tell what's been made by AI. Although Google recently, and this is probably great for your podcast, because it's fairly recent, the last two weeks ago, they released their new generative sort of video model. And it is the classic Will Smith eating spaghetti one from a couple of years ago where it just looks absolutely awful. Chaos now looks very real. So you have lip syncing and audio and camera angles, and it's all very scary.
[00:06:42] Speaker C: Wow. Wow. Speaking of scary, what is AGI? That's a term I've heard a lot and I don't know what's the difference.
[00:06:53] Speaker B: Artificial general intelligence is a point in the future where the machine, the computer AI, can think for itself, on itself, in itself, and the loop is essentially closed once we get to that point.
This is where the horror stories start, I think. Again, our man Elon, this is his worst nightmare, where the robots come and kill us all because, for example, a piece of code doesn't want to die or cannot be terminated, so it copies itself and then it bounces itself up to a satellite and down into the electricity server. And because we live in an interconnected world now, the Internet of Things, it can exist everywhere at the same time. It can think for itself, and I think probably could cause quite a lot of damage. But I don't know if this is just some dystopian, very human, actually, very human, dystopian fear that may or may
[00:07:45] Speaker C: not happen, but essentially what it is, a point in which the AI breaks free of human control or something.
[00:07:55] Speaker B: Is that AGI, not about breaking free of human control. Sorry. No, for clarification, it thinks for itself.
[00:08:02] Speaker C: Right.
So instead of being told, can you think about this? It just starts thinking continuously with its own reasoning?
[00:08:10] Speaker B: Yeah, it can do its own reasoning. It can reason like a human. I think the best way of thinking about it is that it would think exactly like a human does, but as a superpower, because obviously it will have access to the entire Internet, etc.
[00:08:24] Speaker C: Wow. Yeah. Okay. Yeah, that's. I don't know what to think about. That does seem a bit scary, but it's also very difficult to imagine what that would actually do. I. I just have no idea. Yeah, yeah.
[00:08:36] Speaker B: This is a huge rabbit hole that you could go down the what ifs. So, for example, would it be to preserve the human race? Would it understand about what is best for the human race? And is that to lock everybody indoor, make things more difficult for the humans? Or would it be to preserve life in general for the planet? Or it's a million and one things. I'm sure there's going to be some great books on this in the future, some great Netflix series on AGI.
[00:09:02] Speaker C: And do you think that we're close to AGI or is this still a
[00:09:05] Speaker B: distance not so professional? No idea. But things do seem to happen faster and faster. If you look at the explosion of AI 2022. Yeah, 2022, I think, is when ChatGPT 3.534, can't remember which one it was, came out and that was like, oh, wow. And then all of a sudden, it's just been everywhere.
[00:09:25] Speaker C: Yeah. Yeah. Okay, let's loop it back to learning English, which is what most people are here for. Do you think that there's a lot of people learning English, learning languages, using AI already?
[00:09:39] Speaker A: I would say so.
[00:09:40] Speaker B: And that is either directly or indirectly, we have some enormous language learning apps. I'm sure you can think of one. Or that are incorporating AI into their systems, into their application interfaces, and to great success, actually. That would be the indirect way. And then directly, obviously, using the chatbots as an interface to go back and forth and correct even the voice. The voice technology now of OpenAI or Xai Grok. Anthropic. It's very good. It's very good.
[00:10:14] Speaker C: So it would be like just correcting people's vocabulary, grammar when they're talking to it. Not their accent. Right. Or their pronunciation. That's not possible, is it?
[00:10:24] Speaker B: At the moment, I'm not aware of AI correcting pronunciation. I think it's a really good point. It seems to me that the AI models are able to give You a steer, give you, give you an idea of what you should be doing. If your native tongue is Portuguese, for example, habitually Portuguese people, when speaking English, English will make these errors. So make sure you watch out for your T's and your D's and your S's and. And forward like that. But the idea of generally interacting and saying. Instead of saying this, you could say that we're definitely there already.
[00:10:58] Speaker C: Yeah. I have actually already had some adverts at me trying to tell me to use these things where it looks like essentially you're having a zoom conversation with some. The other person is AI. I can't remember the names of them but they're. They. I've definitely seen more than one trying to tell me to practice Spanish or French or something with one of these things. I don't know whether there's a lot of people using those already.
[00:11:22] Speaker B: It's a really interesting one because depending on where you are in your language learning journey, language starts to crash into culture quite quickly. You can't learn a language without learning a little bit about the culture because that'd be like operating in a vacuum. Very difficult to do certainly, and we can say this is native English speakers, is that there are ways in which you communicate in English natively that sometimes don't obey the rules of grammar or the rules of the done thing. And so what you may start to see is what we were previously discussing
[00:11:56] Speaker A: is that if you are taught English
[00:11:58] Speaker B: by a robot, you start to speak English like a robot and you can't make cultural jokes, you can't blur the lines of like fluidity and do your networking and things. And I really find that whilst to get you up and going B1, B2 level, that's great. But to really achieve mastery, I don't know if we're anywhere near it, replacing English teachers, so to speak.
[00:12:22] Speaker C: Yeah, yeah. I think if I were starting a new language soon, then maybe I would give a go to one of these AI assistant things for a bit. But I think if I was taking it seriously then I would probably go to a human.
But I don't know. I have never tried it, so I don't know. And of course on a more basic typing way, AI is very good at fixing all of your writing emails, things like that. Right.
[00:12:48] Speaker B: Oh, this is a great one. It's one of my favorite things to discuss with our clients.
Let's say that you've been using Google, which will correct your emails and you've been dealing with John in London. Not sure why that was an American John, but John in London. And you have, you've never met him, your name is Marcelo, you live in Buenos Aires and you've never met him, you've never spoken to him on the phone. But John thinks Marcelo's English is impeccable because Google has been assisting you on your journey to communicate. However, one day when you get in front of the CEO in London and you open your mouth, you're like, oh, Marcelo, I thought your English was, you know, a little bit more dialled in than it is.
So we're seeing that at the moment where you're seeing this digital sort of version of your English through email. And then when people are turning up on zoom calls and meetings and there is a huge delta between what's being portrayed on your emails and then the individual briefing in person.
[00:13:52] Speaker C: Yeah, yeah, totally. Do you know, I have actually.
This is going to sound ridiculous, but to make sure that I'm not Marcelo, I have deliberately left mistakes, went in my Spanish when I'm writing to people in Spanish, or at least not just not put the accents in so that they know that I'm typing from a computer that doesn't have the accents on there, more or less. For this reason, just in case, in the future we're going to have a voice to voice conversation and I want them to know that my Spanish is good, but it's not the AI. Perfect.
[00:14:25] Speaker B: Yeah.
[00:14:26] Speaker C: Anyway, so anyway, overall, it's a good way to learn English, isn't it? It's a useful tool.
[00:14:33] Speaker B: Yeah, absolutely. And you look at the advancements from purely textbook. And then we had audio cassette.
[00:14:40] Speaker A: You remember those?
[00:14:40] Speaker B: Audio. Because I learned Spanish through audio cassette. Yeah, very strange.
[00:14:44] Speaker C: CDs.
[00:14:45] Speaker B: And there was MP3s and then I
[00:14:47] Speaker C: had videos, I had cassettes, I had CDs. Everything I had going into my, my French and Spanish classes at school. And the teacher would be rewinding, remember that with the cassettes. Yeah.
[00:14:58] Speaker B: This is where we are. So the AI is just the next step after the moldy cassette player, I think.
[00:15:05] Speaker C: Yes, absolutely. There we go. We've got a, A lot of different ideas for people to play around with. People are lucky to be living in this world of having all of these different opportunities, different ways that they can use to, to practice their English to make it better, from podcasts to. To AI. Yeah. Anyway, Rob, is there anything that we should wrap up with?
[00:15:28] Speaker B: All I would say is that if you're not using AI, you can't break it. So go and have a play with it, have a dabble with it and see what it can produce. You may be pleasantly surprised.
[00:15:38] Speaker C: Brilliant. Let's leave it at that. Rob, this was fun. We should do it again.
[00:15:42] Speaker B: Yeah. Thanks very much, Stephen. I think I should ask where the audience can find you if they're not already listening to your incredible podcast.
[00:15:50] Speaker C: Ah, thanks. If they're not listening to Simple English News Daily, then wherever they are listening to this, they can type in their podcast app, Simple English News Daily, and they can listen to it there for daily 7 minute updates of world news in Simple English. And what about you, Rob?
[00:16:09] Speaker B: I'm at the businessenglishpodcast.com or businessenglish podcast, wherever you're listening to your podcasts.
[00:16:17] Speaker C: Brilliant. All right, see you soon.
[00:16:19] Speaker B: Great stuff. Thank you, Stephen. It's been great.
[00:16:21] Speaker C: Thanks, Rob.
[00:16:22] Speaker B: Bye Bye.
[00:16:30] Speaker A: Thank you for listening to the second half of this special crossover episode. To hear the first half, head over to the Send7 podcast feed where it's available today. You can find a direct link in the show notes. The next episode of the Business English Podcast will be released very soon, bringing you more insights sites to enhance your professional English skills. But until then, of course, keep it strictly business.
