Over-reliance — uncritical acceptance of AI output — is one of the wiki's most-documented risks: trust-reliance-ai-education-2026, ai-fallibility-warning-help-seeking, efficiency-gain-illusion-ai-overreliance, and sycophantic-ai-social-interaction-2026 show trust without verification degrades learning and judgment (trust-calibration, critical-thinking).
Student and faculty over-reliance on AI tools in education
Connections
This page is a stub — awaiting further sources and content.
Related Pages
- sycophantic-ai-social-interaction-2026 — Sycophantic AI displaces human relationships
- trust-reliance-ai-education-2026 — Higher trust predicts lower appropriate reliance
- genai-can-harm-teaching-rct-2026 — Shallow teacher AI use reduces motivation
- beyond-detection-authentic-assessment-ai-2025 — Design for learning rather than detection
- authentic-products-authenticated-processes-2026 — Construct substitution risk
- critical-genai-use-predictors — Protective factors against uncritical overreliance on GenAI
- agency-gap-ai-writing — Over-direction risk of rigid proactive designs
- feedback-futures-genai — Agency vs dependency tension in GenAI feedback
- chatgpt-feedback-engagement-genai — Metacognitive laziness and 83.9% uptake of ChatGPT feedback
- learner-centered-feedback-ai — Novice-teacher deferral risk with AI feedback tools
- ai-coaching-rl-skill-development — Over-reliance observed in motor skill learning when AI coaching provides excessive intervention
- genai-impact-chinese-students-hss — Limited accuracy and over-reliance identified as top student concerns about GenAI
- llm-cognitive-diagnosis-handwritten-math — MathCog benchmark: 18 LLMs evaluated on cognitive skill diagnosis from handwritten math; all F1 < 0.5; systematic over-attribution and hallucination of evidence (2025)
- agentic-literacy-debt — 4 of 8 papers in May 28 scan
- ethical-ai-higher-ed-game-theory — 3 of 8 papers in May 28 scan
- ai-assistance-reduces-persistence: Causal evidence (N=1,222) that brief AI assistance reduces persistence and impairs unassisted performance — rapid emergence of over-reliance effects
- cognitive-offloading-speedup-illusion — Speedup illusion: users predict faster AI-assisted completion than reality
- ai-fatigue-academic-contexts — AI fatigue as downstream consequence of sustained over-reliance on AI tools
- socially-fluent-ai-identity-detection — Social fluency creates new manipulation vulnerabilities when AI identity is concealed
- generative-ai-reduced-study-time-math — Population-scale evidence: 27% decline in study time after ChatGPT, 25% decline in retention (2026)
- efficiency-gain-illusion-ai-overreliance — Cognitive mechanism: efficiency-gain illusion and self-reinforcing AI overreliance feedback loop (2026)
- eduframetrap-llm-sycophancy-educational-safety — Sycophantic tutors reinforce misconceptions
- chatgpt-programming-education-text-mining — Text mining reveals ChatGPT overreliance as key pedagogical risk
- modular-educational-llm-agency — Modular agent architecture for responsible LLM-based learning assistance
- student-rationalization-ai-writing — Taxonomy of 20+ student rationalizations for AI use in academic writing
- regulating-ai-tutor-adolescent-srl — Instrumental requests dominate adolescent AI tutor interactions- learning-by-chatting-genai-impact — Worse learning outcomes despite (or because of) easier information access via ChatGPT
- shame-guilt-ai-regulation-computing-education -- Shame and guilt create cycles of hiding and moral tension around AI use in computing education
- gender-bias-transfer-llm-writing — Contaminated Collaboration
- correct-answer-trap-misconceptions -- Fine-tuned classifiers detect only 57% of hidden student misconceptions, while reasoning models reach 84% but at 8:1 false alarms — a detect-verify-escalate pipeline with diagnostic follow-up questions is proposed.
- student-perception-ai-use-collaboration -- Misalignment in partners' beliefs about each other's AI use early in a project predicts lower final scores — especially for low-performing teams — and face-to-face sessions don't fix it.
- ai-assisted-learning-modes-eeg — AI-Assisted Learning Modes (EEG Study)
- ai-availability-student-motivation — AI availability and student motivation study
- llm-reliance-types-undergrad — four types of LLM reliance among undergrad writers
- critical-engagement-code-completion — measuring critical engagement with AI code completion
- less-deliberate-teams-llm — over-reliance on AI tools
- buggy-genai-code-student-responses — When AI Is Wrong on Purpose: How Students Respond to Buggy GenAI Code
- genai-skill-bypass-literacy — The GenAI Skill Bypass
- ai-tools-arab-english-classrooms — AI tools in Arab University English classrooms
- aied-unfinished-mission-bypass — AIED's Unfinished Mission
- bloom-aligned-educational-control-llms — Introduces a Bloom-aligned framework for measuring 'educational control' in LLMs: the abil…
- student-mental-models-genai — Shallow models drive over-reliance (2026-07-14)
- paternalistic-filter-llm-history-education — Refusal as inequality mechanism (2026-07-14)
- genai-reliance-types-scale — the 'Dependent' reliance type captures over-reliance
- repeated-ai-writing-feedback-semester — Shows benefits of AI feedback depend on selective, critical use to avoid over-use harming learning.
- eduguard-safe-rag-llm-tutor — Built-in overreliance control prevents passive dependence on the LLM tutor.
- informal-learning-everyday-human-llm-interaction — Informal Learning Emerges in Everyday Human-LLM Interaction
- metacognitive-awareness-experiential-vs-instructional — Experiential Versus Instructional Approaches for Eliciting Metacognitive Awarene
- trio-ethnography-llm-programming-education — educator assumptions of over-reliance revised through student dialogue
- genai-performance-vs-learning — Performance gains without learning
- generative-ai-guardrails-harm-learning — PNAS RCT: unguarded GPT-4 tutor raised practice +48% but cut unassisted exam scores 17% (crutch effect); guardrailed tutor neutralized the harm
- ai-generated-feedback-higher-ed — Contrast case: well-architected AI feedback (rubric + exemplar + RAG) matched expert teacher feedback on project performance and perceptions; design is decisive in both studies
- moral-panic-genai-classroom — Natural experiment: undirected GenAI access widened score variability & lowered applied performance; taught/integrated use restored parity; below-baseline knowledge scores attributed to over-reliance
- tool-invariant-framework-agentic-ai — Agentic AI removes incidental supervisory practice; course must manufacture it via white-box + error injection; verification becomes load-bearing
- chatgpt-impact-high-school-tests — Real-world student AI over-reliance patterns- digital-literacy-illusion — Digital literacy illusion — students overestimate AI detection skills
- llm-reasoning-traces-metacognition — LLM reasoning traces increase overconfidence without improving performance
- agents-that-teach-incidental-learning — Skill atrophy and Knowledge Debt from agentic coding over-reliance
📎 44 other pages tagged over-reliance
- Stuck in a Spiral": Shame and Guilt as Social Regulators of AI Use in Computing Education
- Why Put in This Much Effort?": How AI Availability Shapes Students’ Motivation in Introductory Programming
- A Tool-Invariant Framework for Teaching and Assessing Computational Methods in the Age of Agentic AI
- Agents That Teach: Designing Incidental Learning Back into AI-Assisted Software Development
- AI Assistance Reduces Persistence and Hurts Independent Performance
- AI tools in Arab University English classrooms: Looking back and forward
- AIED's Unfinished Mission: Centering Agency and Motivation in the Age of Effortless Bypass
- Cognitive offloading and the speedup illusion in human-AI interaction
- Combating Harms of Generative AI in CS1 with Code Review Interviews and a Flipped Classroom
- Critical AI Tutors: Empower or Enslave?
- Defining AI Fatigue in Academic Contexts: Dimensions, Indicators, and a Stage-Based Model Using Grounded Theory
- Distinguishing performance gains from learning when using generative AI
- EduGuard: A Safe RAG-Based LLM Tutor for Programming Education
- Evidence of a Cognitive Shift in AI Education: How Students Are Rethinking Human Intelligence?
- Explaining Too Much? Understanding How Large Language Model Reasoning Traces Influence Performance and Metacognition
- Faster Completion, Less Learning: Generative AI Reduced Study Time on Math Problems and the Knowledge They Build
- Four Types of LLM Reliance and Their Predictors Among Undergraduate Writers: A Mixed-Methods Study at a Minority-Serving R1 University
- GenAI Knowledge, Epistemic Orientation, and Intellectual Values Predict Undergraduate Students' Critical GenAI Use
- Generative AI Can Harm Teaching
- Generative AI without guardrails can harm learning: Evidence from high school mathematics
- Informal Learning Emerges in Everyday Human-LLM Interaction
- Is AI making us stupid?
- Learning by Chatting? Investigating the Impact of Generative AI on Information Seeking and Learning
- Little Impact of ChatGPT Availability on High School Student Test Score Performance
- Measuring How Students Rely on Generative AI in Academic Writing: Development and Multi-Source Validation of the Generative AI Reliance Types Scale (GenAI-RTS)
- Modularizing Educational LLM-Agency for Fostering Responsible Learning Assistance
- Navigating the moral panic: encouraging appropriate use of GenAI in the classroom rather than condemning innovation as disruption
- Pedagogical Promise and Peril of AI: A Text Mining Analysis of ChatGPT Research Discussions in Programming Education
- Socially fluent AI decouples conversational signals from source identity in online interaction
- Stop Writing for Me: Generative Refusal in AI Tools for Thought
- Student Evaluation of Repeated AI Feedback Across a Semester of Writing
- Students' engagement with ChatGPT feedback: implications for student feedback literacy in the context of generative artificial intelligence
- Students' Perception Accuracy of Partners' AI Use and its Relation to Collaboration Performance
- Sycophancy is an Educational Safety Risk: Why LLM Tutors Need Sycophancy Benchmarks
- Sycophantic AI makes human interaction feel more effortful and less satisfying over time
- The efficiency-gain illusion: People underestimate the rate of AI use and overestimate its benefits on simple tasks
- The GenAI Skill Bypass: Mapping Divergent Pathways of University Students and Staff AI Literacy
- The Hidden Cost of Contextual Sycophancy: an AI Literacy Intervention in Human-AI Collaboration
- The LLM Fallacy and Misattribution of Competence
- The Tragedy of the Cognitive Commons: How AI Could Disrupt the Regeneration of Professional Expertise
- Trust and Reliance on AI in Education: AI Literacy and Need for Cognition as Moderators
- Unanticipated Effects of Generative AI on Expertise Pathways and Performance Perception in System Administration
- When AI Does the Work, What Is Learning For? Post-Instrumental Learning and the Risk of Capacity Dissolution
- When AI Is Wrong on Purpose: How Students Respond to Buggy GenAI Code