Hi Everyone,
This is Kyle and Rachel’s contribution to the Mobile feast. We looked at the analog past, mobile present and possible AI driven futures for Language Learning Assessment.
By focusing on several popular language learning applications, it became very clear that AI is driving remarkable change in the industry. If only 25% of the ideas being tossed around right now work out we can expect a lot of fascinating changes to come in the near future.
We look forward to hearing your thoughts and opinions! Below you will find our discussion questions.
https://sites.google.com/view/523-a3-ai-assessment/home
Authenticity vs. Convenience: Do you think apps like Duolingo, Rosetta Stone, and Babbel assess real language ability or mostly app-specific performance? How should educators weigh that difference?
Feedback Quality: The apps all provide instant correction, but is “right/wrong + retry” enough to build deeper skills? What would authentic feedback look like on mobile?
Equity and Access: These apps claim accessibility, but features like pronunciation scoring or AI chats often sit behind paywalls. Does this reinforce inequities in language learning, or does the free tier still make a meaningful difference?
Teacher’s Role: If learners are practicing daily with mobile assessments, what role should teachers/tutors play? Do they become mentors checking process or evaluators verifying proficiency?
Hey Kyle, Rachel,
I must commend you on the flow and layout of your Google site; it was a pleasure to navigate through and helped my own cognitive load focus on the ideas you presented me instead of how to find my way through it.
I’m glad you mentioned the revival of Indigenous languages through the help of AI; I’m using a similar topic in one of the case studies I’m building for my own group’s OER.
I also appreciate you mentioning the over-reliance on AI feedback without actually considering why something might be wrong or out of place. I think part of the joy of learning a new language is getting to understand the nuance and life of the language and that can be hard to do without really studying it and considering how its people use its syntax and turns of phrase to make meaning.
What do you think could be lost from languages if people were to continue to learn them solely through the use of AI? Would someone who learned Spanish, for example, from an AI companion have a more “hollow” grasp of the language than someone who learned it traditionally? If so, what aspects of a language do you think are lost first? Is there a way to preserve those parts of language?
Wonderful module team. I specifically enjoyed the contrast of ‘bad AI’ to ‘good AI’. As with your Duolingo case study, I have become quite skeptical of large tech companies and a turn to AI cost-reduction strategies.
But you highlights of the good AI can bring in preserving and relearning languages is a promising advancement.
Hi Kyle and Rachel,
Thanks for sharing this thoughtful exploration of how AI is shaping language learning assessments. I really appreciate how you connected the analog past, mobile present, and AI-driven future—it gives a clear sense of how fast things are changing.
I think the teacher’s role is shifting from just grading quizzes to becoming more of a mentor and cultural guide. Since mobile assessments can already handle repetitive drills and instant feedback, teachers don’t need to repeat that process. Instead, they can help students set goals, reflect on their progress, and make sense of AI feedback. Teachers can also design authentic tasks like conversations, peer reviews, and creative projects that connect learning to real communication and culture. They still play an important role in verifying the kinds of skills AI can’t measure—like nuance, empathy, and contextual understanding.
In this way, AI supports daily practice, but teachers keep the learning human, reflective, and meaningful.
Inspired by Charlie Gedeon’s TEDx talk “Is AI making us dumber? Maybe,” (https://www.youtube.com/watch?v=m8WomdCLBqE) how can teachers design AI-supported assessments that promote productive resistance—encouraging students to think critically and struggle productively—instead of simply making learning faster or easier?
I really enjoyed the way you broke down the progression of language assessments because it shows a shift in education pedagogy related to language acquisition (i.e., recall to mastery).
While I was teaching math, our school invested in the IXL program for supplementary practice at home. The idea behind this is that students would be able to work on skills and concepts to meet their individual needs. Students commented about how they were achieving ‘mastery’ using the program, but were still unable to apply and complete higher-level thinking questions at school. I am not familiar with the language-based apps, but IXL did provide a diagnostic test and tracked the progress for each student. Moreover, because it was voluntarily used, few students took advantage of using the program.
While mobile apps provide additional means of practice, I think they cannot take away from the teacher’s instruction and ability to diagnose where students might struggle (i.e., whether they missed a skill/concept in elementary school, so are struggling with grade-level content). With the power of public perception, I think the role of the teacher in the classroom has to be clearly outlined, especially if we are expecting daily mobile use.
Man, I remember stumbling across IXL my first year teaching and thought it was a great resource for giving students additional practice in school for Math. Haven’t heard a lot about it since then, though, and once I moved to high school I stopped using it. It was nice to have a repository of skill drills aligned with the BC curriculum.
This is a great OER! I also like how most of these assessments are low-stakes, ungraded assessment. I also observe that these assessments are making use of “Doer” Effect where promote retrieval practice by requiring learners to actively recall and apply knowledge during the learning process (Koedinger et al., 2016).
Personally, I believe that even in the age of AI and mobile intelligence, the role of educators remains essential. While AI can support assessment and feedback, educators still serve as subject matter experts who make pedagogical judgments to guide and deepen students’ learning journeys.
Reference: Koedinger, K. R., McLaughlin, E. A., Jia, J. Z., & Bier, N. L. (2016). Is the doer effect a causal relationship? How can we tell and why it’s important. Proceedings of the Sixth International Conference on Learning Analytics & Knowledge, 388–397. https://doi.org/10.1145/2883851.2883957
I believe apps like Duolingo are excellent tools for helping users learn the foundational concepts of a language, such as vocabulary, grammar, tenses, and basic sentence structures. However, they are less effective for assessing true language proficiency, which develops primarily through active use in conversation and real-world contexts. While these apps provide structured practice and reinforcement of foundational skills, authentic language use, including nuances, language-specific phrases, slang, and conversational speaking, can only be learned through actual communication and social interaction in the studied language. As an English teacher, I see a similar principle in the classroom. I may give students exercises focused on grammar, spelling, or literary devices, but the real measure of their learning comes from how they can apply these skills. Memorizing rules or definitions is only the first step; true understanding is demonstrated when students can transfer their knowledge across contexts, whether through writing, discussion, analysis, or creative projects. Similarly, language-learning apps are valuable for building foundational skills, but authentic proficiency is developed through application, interaction, and contexts beyond the exercises. For teachers, these apps can help assess foundational understanding, while conversational and applied skills require complementary activities.