Rethinking Education in the Age of AI

March 2023--ChatGPT Leave a comment

Rethinking Education in the Age of AI

By: Louise Ohashi

“Alarmed by A.I. Chatbots, Universities Start Revamping How They Teach”

Startling headlines like this from the New York Times (Huang, 2023, January 16)[1] have been widespread since ChatGPT’s release, but how is the average teacher responding? I’m currently analysing data from a worldwide study with 368 language teachers to find out (check Twitter for updates), but in the meantime a quick search online shows reactions at both ends of the spectrum. On one side, innovators have already created learning materials and set AI-integrated assignments. On the other, ChatGPT has been banned and all written work moved to class time to prevent cheating. Love it or hate it, ChatGPT cannot be ignored because we are in a new Age of AI. The dial is about to be turned up.

[1] This article was reproduced in The Japan Times two days later.

For some, the outlook for language education is bleak. An opinion piece on AI (ChatGPT, DALL-E image generator, and machine translation) by Yong Zhao, Professor in Educational Leadership at Melbourne University, claimed: “If machines can do better than most human foreign language learners in communicating in another language, perhaps it is time to rethink traditional foreign language courses” (2023, January 31). I agree with him on that, but we certainly diverge on what he wrote next:

Culture should be the core content of foreign language teaching, but unfortunately linguistic competence has instead always been the primary focus. Assessment in foreign languages has always been on the mastering of vocabulary, grammar, and abilities in reading, writing, listening, and speaking. Given the difficult[y] in achieving linguistic competence in foreign languages and the availability of increasingly sophisticated machine translation, it seems reasonable to rethink foreign language courses as primarily cultural courses, with limited expectations of linguistic proficiency in the target language.

At a basic level, this suggests that foreign language education is failing, so it’s time to focus on inter-cultural understanding instead and outsource foreign language communication to machines. To conjure up an image of the world such a shift could create, I used OpenAI’s image generator, DALL-E 2, guiding it with the words: “Show a business meeting with an equal number of men and women. They should all have cell phones in their hands.” Will international business in 2030 look like this? I’d be happy to see this increase in female representation, but making machines so central to communication would rob us of something distinctly human. During the pandemic, communicating online became the norm. Now that we’re returning to shared spaces—be it for business or pleasure—we should use our own voices and minds, not machines.

Even if AI can communicate on our behalf, that doesn’t mean it should. On the contrary, educators and language learners should use it to develop language skills rather than erasing the need for them. I’m a daily user of machine translation in multiple languages and can testify to its learning potential when used selectively. Furthermore, my research with 153 foreign language teachers at Japanese universities showed the majority agreed with using machine translation as a learning tool, despite expressing concerns over its potential misuse (Ohashi, 2022). The key point, both with machine translation and the newest wave of AI, is knowing how to use it well and when to put it down and think for yourself.

Ideas for using ChatGPT in language education have flooded social and mainstream media, and can be found in other parts of this special issue, so instead of sharing more here I’ll draw attention to three central concepts: the benefit of having “someone” to ask for guidance; the increased value of creativity; and the vital role of critical thinking skills.

“Someone” To Guide Learning

Consulting ChatGPT for guidance can help us to reach new heights. As early adopter Ethan Mollick (2022, December 14) wrote in the Harvard Business Review:

I introduced my undergraduate entrepreneurship students to the new AI system, and before I was done talking, one of my students had used it to create the code for a startup prototype using code libraries they had never seen before. They completed a four-hour project in less than an hour.

This shows how AI allowed students to reach beyond their ability level, moving past what they couldn’t do alone to achieve what could only be done with help. If thoughts of Vygotsky’s Zone of Proximal Development (ZPD) are flashing through your mind, you are not alone. Vygotsky defined the ZPD as:

The distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers. (1978, p. 86)

In language education, teachers are key providers of Vygotsky’s “guidance,” but AI can and will start playing a greater role. This guidance comes not from an adult, not from a peer, but from an artificial “someone” that is always there to help. If language learners don’t understand a verb tense and want explanations, example sentences, and practice tasks, they only need to ask. When they want new reading materials on topics that interest them, “someone” is there to write a story and can modify its linguistic complexity if instructed. Learners can ask without embarrassment, because this “someone” never judges. To get the right help, though, teachers and students must know what and how to ask–practicing creativity–and when to question the output–practicing critical thinking.

The Centrality of Creativity

In the most watched TED Talk of all time (almost 75 million views), “Do Schools Kill Creativity?”, Ken Robinson said: “My contention is that creativity now is as important in education as literacy, and we should treat it with the same status” (2006, June). As ChatGPT and rival AI services (e.g., Google Bard) become commonplace, teachers and students alike will need to modify their way of thinking about education. What do teachers want students to learn? What value is there in the product and the process? What paths can they take to get there? In the short time since ChatGPT’s release, users have flooded the internet with accounts of their experiences. Those who have achieved successful outcomes have contemplated ChatGTP’s affordances and considered how to overcome its limitations. Detailed, specific instructions guide ChatGPT in the right direction. In short, creative minds find the right commands and imagine innovative tasks. If users cannot think creatively, they will not fully harness the potential of AI.

The Cruciality of Critical Thinking

Critical thinking skills play a crucial role in selecting input and assessing output. The potential of AI varies considerably depending on who is commanding it, and as ChatGPT is not always factually accurate, misunderstanding and misinformation will abound if output is uncritically swallowed whole. For instance, it can create an essay and cite references, but in many cases the sources are not related to the content they’re cited for and some are not even real. Teachers and students must make a habit of verifying content externally. This sounds logical given AI’s limitations, but this step is often skipped. Even when the stakes are high, humans are fallible. Perhaps nothing makes this clearer than articles like “Google’s Bard AI Bot Mistake Wipes $100bn Off Shares” (Sharman, 2023, February 9), which reported that Google lost consumer confidence after releasing a promotional video about its new chatbot supplying incorrect information about an easily searchable fact. If company employees don’t bother fact-checking when there is 100 billion dollars at risk, how much can we expect of teachers and students? They are often pressed for time, and it can be tempting to give in and completely outsource tasks to machines instead of putting in the required work. Yet, try we must, and building critical thinking skills into project work and assessment is a step in the right direction.

Ethical Issues

When we view ChatGPT as “someone” who can help and use it creatively and critically, its value cannot be denied. However, there is a hidden price. As pointed out by Uri Gal (2023, February 8), Professor in Business Information Systems at the University of Sydney, nobody gave permission for their personal information and content to be scraped from the internet, and there is no procedure in place for requesting data removal. He also notes that ChatGPT’s data scraping includes copyrighted materials. It’s unclear what repercussions there will be when individuals and companies discover their data and want it deleted or seek compensation. If ChatGPT generates output with copyrighted materials for teachers and students—who may be unaware that copyright was breached—who is at fault? Furthermore, registration for ChatGPT requires an email address and phone number, which limits anonymous use and, as ChatGPT openly states, conversations between users and the chatbot are harvested for training. These privacy issues mean that any student or teacher who uses ChatGPT should do so by choice, not due to institutional requirements.

We must also bear in mind that AI can perpetuate inequality. Did you notice the skin colour in DALL-E 2’s business meeting image? Most AI training is based on data from dominant groups, so ChatGPT is likely to produce output that prioritises them and erases minorities. If we want diversity, we must command it.

Final Thoughts

Teachers will not be made obsolete by ChatGPT, but we need to accept that AI technology is changing the educational landscape. It is vital to consider how to integrate AI in ways that enhance learning and deter misuse. Guiding learners well will require us to draw not only on their creativity and critical thinking skills, but also our own. This doesn’t mean all learning should be AI-assisted. However, closing our eyes and hoping for this stage to pass is not an option, so let’s work together to pave new paths with our AI assistants.

References

Dr. Louise Ohashi (Gakushuin University) specializes in SLA and researches learner autonomy, motivation, and language learning/teaching with technology. She’s Chair of EUROCALL’s AI SIG and an avid language learner who speaks Japanese, Italian, and French (plus her L1, English). She’s also begun her journey with Spanish and German. Twitter: @OhashiLou

 

Return To The Main Issue

Leave a Reply

Your email address will not be published. Required fields are marked *