A survey conducted in 2025 by the Higher Education Policy Institute revealed that 92% of university students now incorporate AI tools in their studies, a significant rise from 66% the previous year, according to Freeman (2025). For educators in online settings, the focus has moved from whether to permit AI to how to develop courses fostering deep learning, irrespective of AI’s involvement. The strategies outlined, derived from current research, classroom experience, and emerging best practices, provide practical methods for integrating AI into online courses while emphasizing genuine learning.
Transitioning from product assessment to process evaluation is crucial. Solely judging students on completed essays, reports, or projects allows AI to perform much of the unnoticed work. Kofinas, Tsay, and Pike (2025) pointed out this issue in a study involving two UK universities, where experienced markers often failed to distinguish student work from AI-generated content, with detection rates dropping to 33%. AI detection tools also pose challenges; Liang et al. (2023) discovered that seven popular detectors incorrectly labeled over 61% of essays by non-native English speakers as AI-generated. Thus, assessment methods should focus on revealing learning progress throughout the process.
In most of my online classes, I use staged assignments with mandatory checkpoints. Students begin by submitting a topic proposal with their personal rationale, followed by an annotated outline with initial sources. They then submit a draft along with a reflective memo explaining their reasoning and any difficulties faced. The final submission includes a revision narrative, creating a documented record of their intellectual growth that AI cannot easily fake. Additionally, students record video walkthroughs of their final projects, explaining their process and challenges. Grading this explanation alongside their work provides a clearer understanding of their comprehension, which is especially beneficial in online courses lacking in-person cues.
Rather than banning AI, it is more effective to assign tasks that require students to use AI tools and critically evaluate the results. In my programming course, for example, students might use AI to explain code or suggest solutions to bugs. They must then test each suggestion, comment on why a fix works or doesn’t, and reflect on what they learned about debugging. I assess their testing process and reasoning, not the AI’s initial suggestions. This promotes deeper engagement than merely copying solutions.
In asynchronous online courses with infrequent instructor interaction, I have introduced a new approach to boost engagement in weekly discussions. Students use AI tools to create practice questions and sample answers, allowing them to self-assess their understanding. They post these AI-generated questions and answers as discussion posts, reflecting on the most helpful questions and identifying gaps in the AI’s knowledge. Additionally, they evaluate at least two question/answer sets from peers, fostering peer dialogue focused on critical assessment and reducing the instructor’s workload in creating quizzes.
Online courses often lead to student isolation, and AI use can exacerbate this if students rely solely on chatbots instead of interacting with peers. Research shows AI tools are most effective in interactive teaching strategies like project-based learning and scaffolded feedback (Long et al. 2026). The technology excels when it complements, rather than replaces, human interaction. Collaborative learning, where students work together towards academic goals, is one of the most effective strategies for enhancing online engagement (Johnson, Johnson, and Smith 2007). In my courses, collaborative activities naturally discourage AI shortcuts.
Integrating AI literacy into the curriculum is essential for responsible AI use. AI awareness modules can provide students with guidelines for appropriate AI use, such as seeking hints versus copying solutions. This framework can be adapted to introductory course modules. Each semester, I conduct an early class session called “AI transparency training,” where students experiment with Generative AI tools on low-stakes assignments, reflecting on the AI’s accuracy and their contributions that AI cannot replicate. This exercise transitions the classroom from secrecy to transparency, setting expectations for the semester.
These strategies emphasize prioritizing the learning process over final products. While AI facilitates polished outputs, it has not replaced the need to assess reasoning, collaboration, and explanation. As Kofinas, Tsay, and Pike (2025) noted, generative AI heightens the importance of social learning, as explicit knowledge can be quickly reproduced by machines. What remains uniquely human is how students apply, question, and communicate their understanding.
The online classroom does not have to be a place where AI lessens learning. Through intentional design, it can become a setting where AI enhances and supports learning. Although this transition requires effort, it leads to a learning experience that is more challenging and better prepares students for the real world they are entering. Taoufik Ennoure, PhD, is a Computer Science Educator and Researcher with over 19 years of experience in Higher Education. He is an Associate Professor at the Community College of Philadelphia and an Adjunct Professor of Computer Science at NYIT, Monroe University, and Baruch College, City University of New York. His research interests include Algorithm Optimization, Artificial Intelligence, Big Data Analytics, and online learning pedagogy.
Original Source: facultyfocus.com
