AI and Critical Thinking in Education

When implemented strategically, AI can transform from a simple answer-generating tool into a sophisticated thinking partner that challenges learners to engage more rigorously with complex ideas and develop stronger reasoning skills.

Several studies indicate that the use of AI tools can enhance critical thinking abilities in specific contexts. For instance, Liu and Wang conducted an intervention study demonstrating that Chinese EFL learners showed significant improvement in critical thinking when using AI tools in literature classes, suggesting these tools facilitate critical cognitive processes through personalized feedback and scaffolding (Liu & Wang, 2024). Similarly, Guo and Wang highlighted that continuous monitoring and real-time feedback from AI systems keep learners engaged and encourage them to consider various perspectives, promoting critical thinking(Guo & Wang, 2024). Moreover, tailored learning experiences enabled by AI have been found to strengthen learners' analytical and problem-solving skills (Walter, 2024).

AI excels at generating thought-provoking questions, counterarguments, and debate prompts that learners can critically evaluate and refine, driving analysis beyond surface-level responses. This approach challenges learners to consider multiple perspectives, identify gaps in reasoning, and strengthen their analytical frameworks through structured intellectual engagement. The key to successful implementation lies in designing AI prompts that progressively increase complexity while requiring learners to justify their reasoning at each step.

• In a history seminar, AI proposes three conflicting causes for the French Revolution; learners evaluate each, identify missing social factors, and refine the list to include class dynamics and economic pressures. This process forces learners to move beyond memorization toward synthesis and critical evaluation of historical evidence.
• In a management course, a custom GPT presents a simulated case study of cross-functional team conflict and prompts, "Should the project manager escalate the issue to senior leadership or implement a team-building intervention first?" Learners critique these options, adjust the debate prompts to explore stakeholder impacts, and deepen their strategic analysis by considering multiple organizational perspectives.

Intelligent tutoring systems can provide hints and challenges tailored to individual learner responses, encouraging justification and revision while maintaining cognitive engagement. This personalized feedback approach ensures that learners remain actively involved in the problem-solving process rather than passively receiving solutions. The adaptive nature of AI feedback allows for differentiated instruction that meets learners at their current level while pushing them toward more sophisticated reasoning.

• In an engineering mechanics course, AI observes a learner's incorrect free-body diagram and prompts, "What force did you omit? Consider the tension in the cable," guiding them to add the missing vector and justify its direction. This targeted intervention maintains the learner's agency in problem-solving while providing just enough support to facilitate learning.
• In a foreign language writing workshop, a custom GPT acting as a writing coach reviews a Spanish essay draft in real time—highlighting misuse of subjunctive verbs, suggesting alternative sentence structures, and challenging the writer to justify their stylistic choices before accepting corrections. This approach develops both language skills and critical evaluation of feedback.

AI's capacity to tailor tasks and examples to individual proficiency levels allows learners to be consistently challenged just beyond their comfort zones while maintaining engagement through relevant, contextualized content. This personalization prevents both boredom from tasks that are too easy and frustration from content that is too difficult, creating optimal conditions for critical thinking development.

• In language learning, the AI assesses a learner's use of past-tense verbs and offers exercises increasingly complex—first regular verbs, then irregular verbs—ensuring mastery before advancing to more sophisticated grammatical structures. This scaffolded progression builds confidence while maintaining appropriate cognitive challenge.
• In a probability & statistics module, the AI begins by asking learners about their interests or hobbies (e.g., "Do you enjoy basketball or video games?"), then generates probability problems using basketball free-throw scenarios or game loot-drop rates, making abstract concepts more relatable and engaging while requiring rigorous mathematical analysis.

AI can facilitate deeper metacognitive engagement by requiring learners to examine not only AI outputs but also their own thinking processes, assumptions, and analytical approaches. This human-in-the-loop verification ensures that learners maintain critical agency while benefiting from AI's capabilities. The reflection component is essential for developing transferable critical thinking skills that extend beyond specific AI interactions.

• Learners critically engage with AI-generated content to check sources, justify arguments, and ensure factual accuracy.
• Learners document their AI interactions by providing transcripts or links and reflect on the experience through a written report.
• Alternatively, learners could create a video presentation, mind map, or another medium of their choice to explain their verification process, illustrate connections to relevant theories or frameworks, and highlight how they identified and addressed any hidden assumptions.