One of AI’s most promising applications in online education is personalised learning. AI-powered adaptive learning platforms analyse student performance data to adjust content delivery in real-time, ensuring that learners receive material suited to their needs and pace (Kumar & Sharma, 2023). Systems like Smart Sparrow and Coursera’s AI-driven recommendations exemplify how AI can tailor educational experiences to individual learners, potentially improving engagement and retention.

AI-driven chatbots and virtual assistants provide students with 24/7 support, answering queries and guiding learners through course materials. For example, Georgia Tech’s AI tutor, Jill Watson, has been used to respond to student inquiries in discussion forums, demonstrating how AI can reduce instructor workload while enhancing student access to immediate feedback (Goel & Polepeddi, 2018). Moreover, AI can identify at-risk students based on engagement patterns, allowing educators to intervene proactively.

Automated grading systems, such as Turnitin’s AI-powered feedback tools, provide rapid, consistent evaluation of student work. AI can assess not only multiple-choice quizzes but also written assignments, offering insights into writing quality, argument structure, and even originality (Brynjolfsson & McAfee, 2022). These technologies hold promise for improving formative assessment in large-scale courses, particularly in MOOCs and distance learning environments.

AI-powered tools assist in creating educational content by generating quizzes, summarising key concepts, and even developing learning materials. AI-assisted instructional design tools, such as OpenAI’s Codex, can support educators in structuring course content more efficiently. However, ensuring the accuracy and pedagogical soundness of AI-generated materials remains an area of concern.

AI models inherit biases from the datasets on which they are trained. If these datasets are unrepresentative, AI systems may reinforce existing inequities, disadvantaging certain student groups (Noble, 2018). For instance, automated grading tools have been found to favour certain linguistic styles, potentially disadvantaging non-native speakers or students from diverse backgrounds.

While AI can assist educators, there is concern that over-reliance on automation may reduce meaningful human interaction in online learning environments. AI should be seen as a complement to, rather than a replacement for, human-led instruction. A balanced approach is needed to ensure that technology enhances, rather than undermines, the role of educators (Selwyn, 2020).

The rise of AI-generated content poses challenges for academic integrity. Tools such as ChatGPT can produce essays and problem solutions, making it difficult to distinguish student work from AI-generated submissions (Cotton et al., 2023). Institutions must develop strategies to address this, such as shifting toward more authentic assessments that are less susceptible to AI assistance.

AI-driven learning platforms rely on extensive data collection, raising concerns about student privacy. Many AI tools track student engagement, interactions, and performance, sometimes without full transparency. Educational institutions must ensure robust data protection policies and ethical AI use in line with GDPR and other regulatory frameworks (Williamson & Eynon, 2022).

AI should be used to enhance teaching, not replace human educators. For example, AI-driven formative feedback can complement instructor guidance, ensuring that students receive well-rounded support while maintaining human oversight.

To mitigate the risks of AI-generated content, educators should implement assessment methods that prioritise higher-order thinking skills, such as project-based learning, oral presentations, and reflective journals. These approaches are less susceptible to automation and encourage genuine student engagement.

Institutions should develop clear policies on AI use in teaching and assessment. Transparency is key—students and educators must understand how AI tools are applied and the implications for learning and academic integrity.