Ceci n'est pas une explication: Evaluating Explanation Failures as Explainability Pitfalls in Language Learning Systems

TL;DR

This paper examines how AI language learning tools can produce misleading explanations that seem helpful but are fundamentally flawed, risking harm and reducing learning effectiveness.

Contribution

It introduces a benchmark for evaluating AI feedback in language learning and analyzes common explanation failures as explainability pitfalls.

Findings

Identifies six key dimensions of effective feedback in language education.

Highlights how AI failures can lead to misconceptions and harm.

Discusses the amplification of risks in language learning contexts.

Abstract

AI-powered language learning tools increasingly provide instant, personalised feedback to millions of learners worldwide. However, this feedback can fail in ways that are difficult for learners--and even teachers--to detect, potentially reinforcing misconceptions and eroding learning outcomes over extended use. We present a portion of L2-Bench, a benchmark for evaluating AI systems in language education that includes (but is not limited to) six critical dimensions of effective feedback: diagnostic accuracy, awareness of appropriacy, causes of error, prioritisation, guidance for improvement, and supporting self-regulation. We analyse how AI systems can fail with respect to these dimensions. These failures, which we argue are conducive to "explainability pitfalls," are AI-generated explanations that appear helpful on the surface but are fundamentally flawed, increasing the risk of attainment, human-AI interaction, and socioaffective harms. We discuss how the specific context of language learning amplifies these risks and outline open questions we believe merit more attention when designing evaluation frameworks specifically. Our analysis aims to expand the community's understanding of both the typology of explainability pitfalls and the contextual dynamics in which they may occur in order to encourage AI developers to better design safe, trustworthy, and effective AI explanations.