Algorithmic drift can cause bias in adaptive tutoring systems
Algorithmic drift in generative adaptive learning
Artificial Intelligence
Summary
Adaptive tutoring systems can unintentionally develop unfair biases over time because small errors in judging students’ abilities grow into bigger gaps. The authors explain this problem as ‘algorithmic drift,’ a feedback loop where mistakes affect future learning paths. They use simulations to show how different learning trackers and curriculum policies contribute to this issue. They also propose a method called the temporal fairness stabilizer, which periodically reassesses learners to reduce unfair achievement gaps better than usual exploration techniques. Their findings highlight fairness as a long-term challenge in educational AI systems.
What this means in practice
- •For adaptive learning platform developers: Implement the temporal fairness stabilizer to periodically reassess learners and reduce bias accumulation in personalized tutoring systems.
- •For educational data engineers: Use agent-based simulations to test how tracking algorithms impact fairness over time in adaptive education systems.
Tested on simulated data.
Authors
Firuz Kamalov, Hana Sulieman, Ahmed El Sayed, Sivaraj R., Nitesh Sughnani
Abstract
Generative artificial intelligence promises to democratize personalized tutoring, yet the longitudinal stability of adaptive systems remains under-theorized. Current ethical discourse focuses on static bias. This paper argues that a distinct threat lies in algorithmic drift: a feedback loop in which minor, demographically asymmetric observation errors compound over time into systemic inequities. We make three contributions. First, we formalize algorithmic drift as a coupled learner--tutor dynamical system. Second, we corroborate this theory with an agent-based Monte Carlo simulation in which either a transparent online tracker or an LSTM deep-knowledge-tracing network drives an IRT-based curriculum policy. Third, we introduce the temporal fairness stabilizer (TFS), a targeted intervention that periodically re-probes learner capability, and benchmark it against standard exploration baselines with explicit group-fairness metrics and replication-level inference. Within the modeled environment, TFS closes the achievement gap where generic exploration does not, and we characterize the conditions under which it succeeds. We separate what is established analytically, what is demonstrated in simulation, and what remains to be validated with real learners. Our results reframe algorithmic fairness in education as a temporal property of a closed loop rather than a deployment-time checkbox.