With the emergence of AI technologies, education is drastically changing. Companies such as Duolingo, an application that allows users to learn new languages, have been quick to implement AI into their organizations. The company uses an AI called the Large Language Model (LLM), which can write prompts for daily learning lessons. The LLM is fed information; from this and previous exercises created by Duolingo experts, it can generate an exercise (Henry, 2024).
With less than 1,000 employees and roughly 21 million daily users, Duolingo needs to prioritize resources to develop its educational platform and meet user demand. Creating and updating classes is a time-consuming task, so the goal of the LLM is to increase overall efficiency.
While many benefits can arise from using AI, such as reducing costs and speeding up the work process, it also raises concern about how this may impact workers at companies shifting towards a higher reliance on these technologies. A study by McKinsey has suggested that by 2030 as much as 14% of the global workforce can expect to change jobs due to AI, that is around 375 million workers (Howarth, 2024). We can already see this impact at Duolingo. In 2023, the company had several waves of layoffs and cut contractors who had previously been working on writing lessons and translating phrases (De Vynck, 2024). Duolingo themselves argued that they are not using AI to replace their workforce, but rather to increase productivity, yet we can see that many workers are losing their job as a result. Their decision to replace human workers with AI-driven content made me think about the true power of AI, its potential is undoubtedly exciting, yet the increasing impact on the workforce, especially in lower-wage areas is unsettling.
Nonetheless, AI skeptics have been quick to point out that GenAI technologies often make errors, but because of the significant cost savings companies are willing to use them for translations, text creation, and design (De Vynck, 2024). A previous Duolingo employee has even suggested that the courses he used to write lessons for have many errors, although a Duolingo representative denied this claim (De Vynck, 2024). It is clear that we are still not at the stage of having AI replace certain jobs completely, but rather using it to simplify and speed up certain tasks. While I often notice the benefits of AI are being presented way more than the risks, I wonder how our workforce will shift in the future. It is already apparent that technology has drastically changed the way we work today, but it raises the question: how positive is AI in the long run and how will it change our job environment?
References
De Vynck, G. (2024, January 10). Duolingo cuts workers as it relies more on AI. The Washington Post. https://www.washingtonpost.com/technology/2024/01/10/duolingo-ai-layoffs/
Henry, P. (2024, May 22). How Duolingo uses AI to create lessons faster. Duolingo Blog. https://blog.duolingo.com/large-language-model-duolingo-lessons/
Howarth, J. (2024, May 27). 60+ stats on AI replacing jobs (2024). Exploding Topics. https://explodingtopics.com/blog/ai-replacing-jobs
Hey, great blog! The way Duolingo is using AI to create lessons is impressive, but it indeed also brings up some concerns. I get how AI can make things more efficient, on the other hand, it is painful to see people lose their jobs, especially when AI is just meant to boost productivity. I wonder how AI will impact other industries, not just education. We’ve seen similar situations where jobs get replaced, and I think it’s important to strike the balance here. AI has great potential, that is for sure, but we need to figure out how to keep people in, especially in tasks where human creativity or judgment plays a big role, like in the entertainment industry for example.