AI in Education – Advancing Learning While Navigating Complex Challenges
- Aug 11, 2025
- 5 min read
Danisa Abiel is well known for her practical solutions to teaching and learning in the advancing fields of Science, Technology, Engineering, and Mathematics (STEM). She is the founder of International Teaching Learning Assessment Consultants and Online Schools (ITLACO). She has authored 20 editions of her newsletter, "The Educator's Diaries," on LinkedIn.
Artificial Intelligence is reshaping classrooms, offering personalised learning, intelligent tutoring, and streamlined assessments that enhance both teaching and student outcomes. Yet, its rapid integration also brings complex challenges around ethics, equity, and educator readiness. This article examines how AI is advancing education, the hurdles it presents, and the steps needed to ensure it complements, not replaces, human connection in learning.
What is AI?
AI stands for Artificial Intelligence. It is a field of computer science focused on creating machines that can perform tasks that typically require human intelligence, such as learning, problem-solving, and decision-making. AI systems use algorithms to process data and make predictions, enabling them to adapt to new inputs and improve their performance over time. The use of AI requires a detailed understanding of how it executes these tasks. For example:
Learning and problem–solving:
AI systems are designed to learn from data and experience, allowing them to adapt to new situations and solve complex problems.
Decision–making:
Many AI applications involve making decisions based on data analysis, such as identifying patterns, predicting outcomes, or recommending actions.
Data–driven:
AI relies heavily on data to train algorithms and improve performance. This data can come from various sources, such as text, images, or sensor inputs.
Various applications:
AI is used in many different fields, including education, healthcare, finance, transportation, and customer service.
Use of AI in education
As education adapts to the rapid pace of a technological world, and student generations become increasingly fast-paced in their learning, AI has emerged as a valuable tool in classrooms and educational institutions. It has been integrated into:
1. Personalised and adaptive learning
AI-driven adaptive learning systems dynamically tailor content and pace based on each student's needs. Extensive research has shown that 86% of studies on adaptive systems report clear positive impacts on student outcomes (SpringerOpen). Platforms like DreamBox Learning and Century Tech offer subject-aligned, interactive pathways that improve engagement and attainment while reducing teacher workload (Third Space Learning). Riiid Labs, Thinkster, and others utilise machine learning to customise curricula and support student performance tracking (Built In).
A notable example is Squirrel AI Learning, which breaks curricula into fine-grained "knowledge points" to target learning gaps precisely, using vast datasets to fine-tune lesson sequences and practice.
2. Intelligent tutoring systems
Intelligent Tutoring Systems (ITS) aim to replicate one-on-one guidance. AutoTutor, for instance, engages students through natural language dialogue, responding to their cognitive and emotional states. Evaluations show substantial learning gains, with average effect sizes around 0.8, comparable to nearly a full letter grade.
Other platforms, such as Learnbot-style tools and StudyBot, exhibit scalable tutoring capacity. The latter is designed to democratise access to personalised guidance via structured pedagogical workflows (Reddit). In a real-world pilot in Nigeria, after-school AI tutoring combined with teacher support yielded gains equivalent to two years of traditional schooling in six weeks, a powerful demonstration of impact in underserved contexts (Reddit).
3. Automated assessment and institutional support
AI tools streamline administrative and evaluative tasks. Research shows that within higher education, most AI applications fall into four categories: personalisation, assessment, profiling/prediction, and tutoring systems (SpringerOpen).
Automated grading, feedback on writing, and content evaluation allow faster instructor responses and better scalability. Some systems generate exam questions via natural language processing, though educators must validate these to maintain alignment and quality (SpringerOpen, Frontiers).
4. Evolving educator roles and pedagogical innovation
AI in education is reshaping the role of instructors, from content deliverers to mentors, data interpreters, and AI literacy facilitators. A significant shift is underway, with teaching becoming learner-centred and empowering student agency through active learning, gamification, and collaborative models (SpringerLink).
However, many educators lack sufficient training in AI integration. Research emphasises the need for professional development and curriculum units on AI literacy to ensure meaningful, ethical, and pedagogically grounded deployment (ResearchGate, SpringerLink).
5. Ethical, social, and practical challenges
Concerns abound regarding bias, privacy, transparency, and the digital divide. Reviews show a lack of theoretical grounding and critical reflection in many AIEd studies, especially regarding ethics and the socio-cultural context of learning (SpringerLink). Risk areas include academic integrity (e.g., plagiarism), algorithmic bias, and data misuse (Frontiers).
Implementation in low-resource or developing contexts may deepen inequity due to infrastructure gaps or algorithmic assumptions tied to global data sets (rsisinternational.org).
Thoughtful frameworks emphasise explainability, transparency, inclusivity, and student and teacher agency, reinforced by governance and evaluation across multiple dimensions of learning outcomes (SpringerOpen).
6. Real world deployments and institutional trends
Khan Academy’s Khanmigo chatbot, built on GPT-4, supports students in math, science, humanities, and essay writing, while also offering tools for teachers. It is now available across dozens of districts and includes teacher-focused generative AI tools offered via a Microsoft partnership (Wikipedia).
Kira Learning, co-founded by Andrew Ng, is expanding from computer science tutoring to general classroom support, with AI agents helping with lesson planning, grading, and student feedback, while freeing educators to mentor effectively (Business Insider).
The American Federation of Teachers, backed by Microsoft, OpenAI, and Anthropic, has launched an AI training initiative to equip educators with practical skills, such as lesson planning and quiz creation, and to guard against technology’s unintended pedagogical consequences (time.com).
Cognitive and social considerations
Recent studies highlight potential drawbacks. An MIT study found that students relying on LLMs like ChatGPT for essay writing showed lower cognitive engagement and diminished neural connectivity compared to those using search or unaided writing, raising concerns about critical-thinking atrophy (theaustralian.com.au).
In the UK, the growing reliance of students on unregulated influencers and AI-generated exam help is raising concerns about misinformation and eroding trust in teaching expertise (thetimes.co.uk).
Conclusion
AI is transforming education by enhancing personalisation, accessibility, and operational efficiency, but its use invites critical challenges around pedagogy, ethics, equity, and human agency. Research underscores that successful implementation only occurs when educators are engaged and equipped, students are empowered with AI literacy, governance is transparent, and tools are rigorously evaluated.
It should be noted that AI is not always "human-like." While AI can mimic some human-like capabilities, it is important to remember that it still operates based on algorithms and data, rather than conscious thought. This leads to other complications that will be discussed in the next edition. It is therefore essential to understand that AI should augment, not replace, the human elements of education.
Read more from Danisa Abiel
References:
Meta‑reviews and systematic studies on AI deployment in education (SpringerLink)
Platform-specific case studies: Squirrel AI, Khanmigo, Kira Learning, DreamBox, Century Tech (Wikipedia)
Studies on ethics, bias, and pedagogy in AI education (arXiv, arXiv, Frontiers, rsisinternational.org, ResearchGate, SpringerOpen)
News articles on policy, training, cognitive impact, and adolescent use of AI (time.com, theaustralian.com.au, AP News, thetimes.co.uk, AP News)
