Article Summary
- AI tools are transforming mathematics learning by shifting students from memorising answers to developing reasoning, reflection, and independent problem-solving when used alongside effective teaching.
- Teachers remain central to the learning process, using AI as a scaffold that encourages explanation, discussion, and deeper conceptual understanding rather than answer dependency.
- Schools such as GIIS Kuala Lumpur are well positioned to embrace this transformation through inquiry-based learning, personalised instruction, and technology-enabled classrooms that encourage critical thinking.
A Malaysian parent recently watched their child solve a difficult algebra problem in seconds using an AI app. The answer was correct, but when asked to explain why it worked, the child struggled. That moment highlights an important question facing families and schools today: Is AI helping students learn mathematics—or simply helping them get answers?
Artificial intelligence is rapidly becoming part of everyday education. The challenge is no longer whether students should use AI, but how they should use it to become stronger thinkers rather than faster answer-seekers.
How is AI changing the way students learn mathematics?
AI is changing mathematics education by supporting students through intelligent tutoring systems, adaptive learning platforms, AI-assisted problem-solving tools, and AI-supported collaborative learning environments.
Rather than replacing traditional teaching, these technologies provide personalised support that responds to each learner’s progress. Intelligent tutoring systems offer hints instead of immediate answers, adaptive platforms adjust question difficulty in real time, AI-powered assistants guide students through complex problems, while collaborative tools encourage discussion and peer learning. Together, these innovations represent one of the most significant developments in Artificial Intelligence in Education.
Research by Aleven et al. (2016) and Holmes et al. (2019) suggests that well-designed AI systems improve learning by scaffolding reasoning rather than simply delivering solutions. Students receive gradual support, reflect on their thinking, and build confidence before progressing independently. Instead of asking, “What’s the answer?”, learners begin asking, “Why does this method work?”
What role do intelligent tutoring systems play in building problem-solving confidence?
Intelligent tutoring systems (ITS) provide personalised guidance throughout the learning process. Instead of correcting mistakes immediately, they offer prompts, hints, and feedback that encourage students to discover solutions independently.
Baker et al. (2020) found that students using tutor-supported Algebra I systems demonstrated higher persistence, stronger problem-solving performance, and greater confidence when tackling unfamiliar mathematical challenges. Because support is delivered only when required, learners gradually develop independence instead of relying on continuous assistance.
What is adaptive learning and how does it help students?
Adaptive learning platforms continuously analyse student performance and adjust lesson difficulty to match individual learning needs. Popular examples include Khan Academy and DreamBox, which personalise practice questions based on previous responses, ensuring students remain challenged without becoming overwhelmed.
Instead of presenting identical exercises to every learner, adaptive systems identify knowledge gaps and provide targeted practice. Students who master concepts quickly move ahead, while those needing additional reinforcement receive further guided activities.
A study by Alqahtani and Mohammad (2021) involving secondary students in Riyadh found that adaptive AI platforms significantly improved strategic thinking, reflective analysis, and self-monitoring behaviours. Students became more aware of their own learning processes while strengthening math problem solving skills, leading to better long-term understanding rather than short-term memorisation.
Can AI replace math teachers in schools?
No. Current research consistently shows that AI works best when guided by skilled teachers rather than replacing them.
AI can personalise instruction, provide immediate feedback, and automate routine practice. However, teachers remain essential for encouraging discussion, asking deeper questions, correcting misconceptions, and helping students connect mathematical concepts to real-world situations.
Gerlich (2025) warns that unrestricted use of answer-generating applications such as Photomath may create “answer dependency,” where students become increasingly reliant on technology instead of reasoning independently. This concern is particularly relevant as more families adopt AI for math applications at home.
Fortunately, research also demonstrates an effective solution. Webel and Otten (2015) showed that teachers successfully transformed Photomath into a learning tool by asking students to explain every step suggested by the application. Rather than accepting answers passively, learners justified each calculation, compared alternative methods, and reflected on underlying mathematical principles.
What happens when students rely on AI without teacher guidance?
Without structured guidance, students may gradually lose confidence in solving unfamiliar problems independently because they become accustomed to immediate solutions.
Research suggests several practical strategies reduce this risk:
- Encourage students to explain each solution in their own words.
- Delay AI hints until learners have attempted the problem independently.
- Ask students to compare multiple solution methods.
- Use AI as feedback after problem-solving rather than before it.
These simple teaching approaches transform AI from an answer provider into a thinking partner.
What learning methods do international schools in Malaysia use to go beyond rote memorisation?
Modern international schools increasingly move beyond memorisation by combining inquiry-based teaching, personalised learning, and carefully guided technology integration.
Research discussed by Alsulami (2016) highlights the limitations of exam-focused, rote learning approaches that prioritise remembering procedures over understanding concepts. In contrast, inquiry-driven classrooms encourage students to investigate patterns, test ideas, discuss reasoning, and solve authentic mathematical problems.
This philosophy is reflected in learning environments such as GIIS Kuala Lumpur, where inquiry-based education, collaborative learning, SMART Campus technology, and personalised instruction encourage students to explore ideas rather than simply memorise formulas. While AI tools may support learning, classroom discussions, teacher guidance, project-based activities, and reflective questioning remain central to developing independent thinkers.
The research also recommends scaffolded problem-solving, where AI assistance gradually reduces as students gain confidence, alongside AI-supported collaborative learning described by Luckin et al. (2016). These approaches help AI for students become a meaningful learning companion instead of a shortcut.
Why do individualised and active learning methods matter in curriculum design?
Students learn mathematics at different speeds and often struggle with different concepts.
Individualised learning allows each learner to receive appropriate levels of challenge, while active learning requires students to explain, analyse, justify, and reflect on mathematical ideas.
The findings from Alqahtani and Mohammad (2021) indicate that adaptive AI environments significantly improve strategic thinking because students actively monitor their understanding instead of passively completing exercises. This personalised, reflective approach supports deeper conceptual learning across diverse classrooms.
What critical thinking skills can AI help students build in mathematics?
When implemented responsibly, AI helps students develop metacognition, self-correction, analytical reasoning, persistence, and reflective decision-making rather than simply improving computational accuracy.
Across the reviewed research, students consistently demonstrated improvements in monitoring their own thinking, recognising errors, evaluating multiple solution pathways, and persisting through challenging tasks before requesting assistance. These habits contribute directly to stronger independent learning.
These findings closely align with Facione’s (2015) framework of Critical Thinking Skills, which emphasises analysis, evaluation, interpretation, inference, explanation, and self-regulation. AI-supported mathematics learning becomes most effective when technology encourages students to practise each of these thinking processes rather than bypass them.
For GIIS Kuala Lumpur, this reinforces the importance of combining digital innovation with teacher-led inquiry. Technology may personalise learning, but meaningful education still depends on students asking thoughtful questions, explaining their reasoning, and developing confidence in solving unfamiliar problems independently.
Traditional vs AI-Assisted Mathematics Learning
| Learning Area | Traditional Learning | AI-Assisted Learning |
| Conceptual understanding | Typically teacher-paced for the whole class | Personalised explanations and adaptive support improve conceptual understanding |
| Analytical problem-solving | Limited individual feedback between assessments | Immediate hints encourage reasoning before revealing solutions |
| Student engagement | Can decline if work is too easy or too difficult | Adaptive difficulty maintains motivation and appropriate challenge |
| Learning persistence | Students may give up after repeated mistakes | Intelligent tutoring systems encourage continued effort through scaffolded support |
| Teacher role | Primary source of instruction | Learning facilitator who guides reflection and discussion |
| Potential challenge | One-size-fits-all instruction | Risk of answer dependency if AI is used without teacher guidance |
Adapted from findings discussed by Baker et al. (2020) and the reviewed research synthesis.
Research Snapshot
Research Basis
- 97 research articles were initially screened.
- 10 high-quality studies met the final inclusion criteria for detailed review.
- The review found consistent evidence that AI-supported mathematics instruction improves engagement, reflective thinking, and problem-solving when combined with effective teacher guidance.
- Alqahtani and Mohammad (2021) also reported significant improvements in student engagement, strategic thinking, and self-monitoring using adaptive AI platforms.
Final Words
Artificial intelligence is reshaping mathematics education, but its greatest value lies not in producing faster answers—it lies in developing stronger thinkers.
The evidence consistently shows that AI delivers the best outcomes when teachers use it to scaffold reasoning, encourage reflection, and promote independent problem-solving. Students become more resilient, more analytical, and more confident because they learn how to think rather than simply what to answer.
At GIIS Kuala Lumpur, this philosophy aligns closely with inquiry-based learning, personalised teaching, collaborative classrooms, and technology-enabled education. By thoughtfully integrating AI into learning experiences while maintaining strong teacher guidance, the school helps students prepare for a future where critical thinking is every bit as valuable as mathematical accuracy.
As AI continues to evolve, the schools that are most likely to succeed will not be those that simply adopt new technology, but those that use it to nurture curious, reflective, and lifelong learners and GIIS KL is one of them. Contact us to learn more!
FAQs
Does using AI tools make students less able to think independently in math?
Not necessarily. Research shows that AI only reduces independent thinking when students rely solely on generated answers. When teachers require explanation, reflection, and justification, AI strengthens independent reasoning instead.
How can schools ensure fair access to AI learning tools for all students?
Schools can provide shared digital resources, structured classroom access, teacher training, and equitable technology infrastructure so every student benefits regardless of their background or home access.
What is the difference between AI-guided tutoring and a traditional math tutor?
AI-guided tutoring provides personalised, instant feedback and adapts continuously to student performance. A traditional tutor contributes human judgement, emotional encouragement, and flexible explanations. Together, they complement each other rather than compete.
How should students use tools like Photomath without losing the learning value?
Students should first attempt problems independently, use AI only after making a genuine effort, carefully review each solution step, and explain the reasoning in their own words. This approach preserves learning while benefiting from AI support.
