Speaker 1: I live in the Tsuen Wan district in Hong Kong, and this is the view that I wake up to every morning. As I prepare to go to work every day, I see thousands of students walking around and running around in the schoolyard, and I cannot help but wonder how the recent technological developments particularly in artificial intelligence are going to impact the lives of thousands of students every day. This is because recently we are going through an artificial intelligence revolution in education. We have GPT-4, we have Palm 2, Microsoft Copilot, Dolly 2, Adobe Firefly, and Mid Journey, and these can do incredible things with text generation, image generation, audio, video, and synthetic data generation. Using GPT-4, for example, you can write a book, for example, in a matter of days, and you can create images. How many of you have a fear that in the future what you are doing now may be replaced by artificial intelligence? And I see some hands. So all those artificial intelligence models, what they can do is to regurgitate whatever there is already on the internet. They cannot be creative, they cannot use common sense. Let me give you an example from Hong Kong. Suppose you are a tourist, you are just visiting Hong Kong, and you are in the Jordan station, which is on the red line over here, and you want to go to Mong Kok East, which is on the blue line there. You ask GPT-4, which MTR line should I take from Jordan to Mong Kok East? It will probably give you two options. One option would be to take the south route, and the other option would be to take the north route. But if you just use your common sense and creativity, you would know that actually, you know what? I don't have to take the train there, I can just walk there. Nevertheless, these language models can do amazing and incredible things. This is some content that I created using an image generator AI, right? I just asked it to create a drawing of an Asian woman in the style of Van Gogh, and this is what I got. And I asked it to create another image of an Asian woman in the style of Salvador Dali, and this is what I got. And you can see some resemblance here, right? To Salvador Dali paintings. The things in the sky, what the woman is holding, the bird. So this is incredible, but it's still an amalgamation of whatever we can find already, whatever already exists on the Internet. Recently, United Nations Educational, Scientific, and Cultural Organization, UNESCO, released a document outlining how GPT-4 particularly can be used in education. Sal Khan recently gave a TED Talk actually outlining how GPT-4 and other large language models can be used. There are so many uses actually, but the two main uses are GPT-4 being used as a personal tutor by each student, and it being used by a teaching assistant by each teacher, right? So teachers can use GPT-4, they can input student work, and they can get personal feedback for each student work. So you don't have to read things anymore if you're a teacher, but we still have to, actually. There are so many other uses. You can use it as a guide, you can use it as a motivator or a study buddy, right? And because GPT-4 and other large language models can be used in education and they have so many uses, this also has implications for work because there are so many examples of AI-exposed work activity. What it means is that there are so many examples of work that can be completed by artificial intelligence tools and that obviates a human being there. So we don't need a human to do those repetitive tasks for us anymore. And because of this reason, Goldman Sachs recently released their report in 2023, and they argue that globally, 18% of work could be automated by artificial intelligence. And very interestingly, the percentage for automation in Hong Kong is about 30%. It's the highest in the world. So it means that a lot of people will be losing their jobs to automation. So then what do we do as educators? Because I can see that some students don't see the value of education anymore. They are questioning why do they even have to attend classes anymore, right? In this talk, I want to argue that there is one and only one thing that we need to do as educators, and that is to make sure that our students are AI-proof. What do I mean by being AI-proof? It's an adjective that means being resilient, being immune so that we will not be replaced by AI in the future. We will not be negatively impacted by artificial intelligence in the future. But being AI-proof also requires us to be human. What do I mean by being human? I mean using and capitalizing on our human skills that cannot be easily replicated by artificial intelligence, right? Now I want to talk about those specific skills. According to the Organization for Economic and Corporation Development, OECD, the most crucial skills that we need to pay attention to in education are cognitive and metacognitive, social and practical skills. The World Economic Forum has a similar take on the skills required for students, and these are cognitive, social and physical skills. And very recently, the World Economic Forum released their Future of Jobs report. They also argue that in the workplace, these are the five skills that the employers seek for the most. And these are analytical thinking, creative thinking, resilience, flexibility and agility, motivation and self-awareness, curiosity and lifelong learning. And I argue that these are the five skills that we also have to try to foster in our classes irrespective of what we are teaching, what content we are teaching. Let's look at a definition for each of those. What is analytical thinking, for example? It's the ability to systematically and logically work through an issue. What about creative thinking? It's finding novel and practical ways to address challenges. What about resilience? It's the ability to recover from setbacks. And I have something special to say about resilience. Sometimes I see among my students that they do not have that much resilience, they don't have that much confidence because they think that their English is not that good enough. Well, let me tell you one thing. Fluency in English is not an indicator of intelligence, and pronunciation in English is not an indicator of intelligence. What about flexibility? It's the ability to change to suit new conditions. Now we have large language models and artificial intelligence. What is next month? What is next year? What awaits us, right? So we have to be ready. Agility is the ability to think and draw conclusions quickly. Motivation is being able to initiate and maintain goal-oriented behaviors. Self-awareness, having conscious knowledge of one's own character and abilities, strengths and weaknesses. Curiosity, we should be eager to learn, we should have an eager desire to learn. And lifelong learning, finally. It's an ongoing voluntary and self-motivated pursuit of knowledge. We have to let our students know that learning is not just limited to classroom environments, learning is a lifelong process, right? So how can we improve all those skills that make us truly human that cannot be easily replicated by artificial intelligence? Well, language learning is one way. For example, I'm Turkish, and I can assure you that if you study Turkish and if you learn Turkish you'll be more analytical. Why? Because you have to analyze such sentences all the time. This is just one word, but in English it's a full sentence. Why? Because in English it means meet, me is a negation, yecek is a future marker, lar is the third person plural, mi is the question, and di is the past tense morpheme. So if you're doing this the whole time while you're learning a language, be it Turkish or another language, of course you'll be more analytical. Let me give you an example from my own Cantonese learning experience. I'm really interested in learning more and more expressions in Cantonese, and recently I came across the saying, I hope you don't mind my horrible Cantonese intonation and tones. Lek gam yat zhong zhi lai mai tin cheung ah. Literally means, thank you so much, literally means you're finally getting married today. So I was thinking about this expression because literally the expression in red means shutting the skylight, so why should it mean to get married? This opened up new horizons for me because we don't have such an expression in English or in Turkish. Let me ask you a question. What is the sound that a cat makes in Cantonese? What about the sound that a dog makes in Cantonese? So let's look cross-linguistically the variation of those nature sounds we call onomatopoeia in linguistics. In English a dog barking is bow wow. In German it's wow wow, French is woof woof, Spanish is wow wow, Hebrew is how how, Hindi is bow bow, Mandarin is won won, Japanese is won won, and Greek is how how. And in a cat meowing, it's kind of similar across the board, it's meow, but interestingly it's Japanese, it's meow, and in Greek it's meow, right, it's more nasal. So learning new languages is a way to understand how individuals with different languages and cultural backgrounds perceive the world differently, and of course learning different languages make us more analytical and creative because it opens up new horizons for us. Let me tell you another way how we can be more analytical and creative, let me give you an example from my own teaching experience. In 2019 I was teaching at Case Western Reserve University in Cleveland, Ohio, and I was supposed to teach a linguistics class. My students didn't seem too interested in linguistics, so I had to motivate them, right? What I did, I said, okay, what we are going to do is we are going to make this class a project-based class, first I'm going to teach you more about linguistics, subfields of linguistics like phonetics, phonology, morphology, syntax and semantics, and pragmatics, and then I'm going to have you create a language from scratch. Students were really puzzled at the idea of creating a language from scratch, they were like, how on earth am I going to create a language? You are crazy, they said. But I gave them an example and I said, you know what, suppose there is a movie, and here is a movie poster, the movie is called Case Closed because it was Case Western Reserve University, and I said, okay, you are supposed to create a language, an alien language for this movie, just like the Navi language in Avatar, or Klingon in Star Trek, or Elvish language in Lord of the Rings, right? And students did amazing work, they used their creativity and they did amazing work. Look at this one. So this student created an alphabet, they created a language and they named it Eagle, and they created alphabets, so these are the international phonetic symbols, and these are the symbols that the student created for their writing systems. So this is a, a, o, i, u, u, a, e, and t, and these are p, b, t, d, you get the idea. And this student came up with a system whereby words appeared in peripheral shapes based on the parts of the speech. For example, verbs appeared in squares, like run, it appears in a square. Nouns and adverbs appeared in triangles, and names, nouns and pronouns appeared in circles. Look at the word ends, because it's a connector conjunction, it doesn't have a peripheral shape, right? So this was kind of fascinating and mind-blowing. And this student came up with some vocabulary items that were really interesting, like pretty but fragile, it's e, u, my, sha, and only in daydreaming or out of reach, pa, in, ku, i, u, and these are the symbols that the student came up with. Yet another student created a language and they called it quat, and this student told me that they were studying chemistry and they were really fascinated by chemical structures and they used those chemical structures to create an orthography, a writing system for their language, and this is what they created. And look at the consonants that they created. Look at p, for example, t, k, f, s, sh, and h. They look very similar, but they are distinct, right? Yet another student created a language and they named it logon, because it's a logogram. A logogram is a language where the symbols represent real-life objects. And the student said, these morphemes are examples of a logogram and resemble the process of raining. For the verb form of rain, o here represents the sun, and this curvy line and the straight line under that represents the cloud, and these vertical lines represent the raindrops. You see how rain as a verb is more dynamic than rain as a noun. That was also very interesting. And this student created some vocabulary items. I want you to look at the word animal and animals. Animal as a singular, it doesn't have a full circle at the bottom. Animals as a plural, it has a full circle at the bottom. So you see the creativity in the language system that the student created, right? And I want you to look at the word animals and flowers, and I want you to look at the word bees. Because animals and flowers, you combine them, you get bees. So in the language that this student created, bees are basically animal flowers, right? Another student came up with a language, they named it Canterin. Actually they didn't create a language, they created a writing system for Cantonese. They were a Cantonese speaker. They told me that they found Cantonese orthography very difficult, traditional Chinese orthography. And they created a new writing system for Cantonese. Look at this. So they said, okay, this could be b, and this could be p, and this could be n, and it goes on like that. And look at these words, for example, lei ho. This is the traditional Chinese orthography, and the student created this, a simplified orthography for Cantonese. And look at joy again, the second one. So at the end of the course, I gave students a questionnaire to see whether they liked the course and whether they were able to master the content. The results show that not only were the students able to master the content of linguistics, but they were also having a lot of fun in the creation of their languages. And the questionnaire also showed that students were able to improve their analytical thinking skills because a lot of students either strongly agreed or agreed with the statement. So why am I talking about learning languages and creating a language as a way to improve our analytical and creative thinking skills and motivation? Why? It's because we are losing linguistic diversity in the world, that's why. And that's a major, major global issue that we are facing. According to UNESCO, there are about 6,000 languages in the world, and we are going to lose about 90% of all those languages in the next 75 years. But languages are a window into human cognition, history, and culture. When we lose a language, we are not just losing a language, we are losing the cognition, culture, and all the traditions that come along with a language. Now I want to ask you, are you AI proof? Because tomorrow is too late to AI proof ourselves. To be AI proof as educators, we have to make sure that we help our students to be AI literate, but we also need to make sure that we help them to capitalize on their core human skills that cannot be replicated by AI. Ending my talk, I want you to complete this sentence. I won't be replaced by technology or AI because... And in completing this sentence, I want you to think of what truly makes you human. Because only by harnessing those skills that truly make us human can we address major global challenges. Thank you.