Critical AI Tutors: Empower or Enslave?

Synthesis: Critical AI Tutors: Empower or Enslave? — A position paper presented at the AIED 2025 workshop that issues a stark warning: unchecked use of AI tutors risks creating a generation of cognitively atrophied learners who have traded genuine understanding for the illusion of competence. Drawing on cognitive science and pedagogical theory, the authors argue that AI tutors without intentional Guardrails lead to cognitive atrophy, loss of agency, emotional risks, and serious ethical concerns around privacy and academic integrity. The paper advocates for critically informed, transparent AI use that empowers rather than diminishes the learner, and calls for student perspectives to be centered in the design and deployment of AI tutoring systems.

Key Findings

This position paper occupies a critical counterpoint in the Intelligent Tutoring literature. While much of the field focuses on optimizing AI tutor performance and learning gains, the authors argue that effectiveness metrics alone are dangerously insufficient — what matters equally are the cognitive and developmental costs that AI tutors may impose on learners.

Cognitive atrophy and the shortcut problem. The central argument draws on Cognitive Load Theory and the well-established finding that effortful cognitive processing is essential for durable learning. When Intelligent Tutoring systems provide immediate, high-quality answers and solutions, they effectively short-circuit the very cognitive processes — struggling with problems, retrieving from memory, constructing explanations — that produce deep understanding. This phenomenon, which the authors term cognitive atrophy, mirrors concerns raised in the The efficiency-gain illusion: People underestimate the rate of AI use and overestimate its benefits on simple tasks and Cognitive Offloading literatures: learners may feel more productive while learning less.

Loss of agency and dependency. Beyond cognitive effects, the paper identifies a broader threat to learner agency. Prolonged reliance on AI tutors can produce Over-Reliance — a state where students lose confidence in their own reasoning abilities and become dependent on AI assistance even for tasks they could complete independently. This dependency dynamic connects to the Catching The Correct Answer Trap: Characterising AI Tutor Blind Spots When Analysing Student Reasoning problem and the finding that Over-Reliance — students give up more quickly when AI help is available.

Emotional and Well-Being risks. The authors highlight underexplored emotional dimensions: AI tutor interactions can erode Self-Efficacy when students compare themselves unfavorably to flawless AI outputs, contribute to Defining AI Fatigue in Academic Contexts: Dimensions, Indicators, and a Stage-Based Model Using Grounded Theory, and diminish the relational aspects of learning that Teaching frameworks seek to preserve.

Ethical concerns. The paper catalogs significant ethical risks including Academic Integrity erosion, questionable Privacy practices in educational AI systems, and the broader societal implications of Is AI making us stupid? — a provocative framing that challenges the edtech optimism narrative.

What this means for practice

• Students. Interrogate every AI answer you receive: ask what it assumes, what it omits, and whether you could have produced it yourself, rather than reading fluency as correctness.
• Students. Protect the effortful part of the work — drafting, retrieving, explaining — and reserve AI help for tasks you can already judge.
• Students. Watch for the dependency signature this paper describes: falling confidence, unfavorable comparison against flawless machine output, and reluctance to attempt work unaided.
• Students. Ask what a tutoring system records and who can see it before you use it, since the paper documents questionable privacy practices in educational AI.

Limitations

• This is a position paper with no study design, participants, or measured outcomes; its claims are arguments from cognitive science and prior literature.
• Student evidence comes from cited surveys — 1,200 young adults aged 18–24 and an international survey of 4,000 university students across 16 countries — not from data the authors collected.
• Constructs such as cognitive atrophy and loss of agency are not operationalized into measurable variables, so the paper cannot estimate the size or speed of the effects it warns about.
• Because it catalogs harms rather than comparing conditions with and without AI tutors, it cannot attribute learning losses to tutor use.

Citation

Favero, L., Pérez-Ortiz, J.-A., Käser, T., & Oliver, N. (2025). Critical AI Tutors: Empower or Enslave?.