Right now, there’s no shortage of positions and views about AI in higher education. You can find articles either warning of the doom AI will bring or praising it as a monumental leap forward for learning, work, and our personal lives. I actually wrote an article sharing my own perspective on AI a few weeks ago, which I published and then promptly removed because I felt I wasn’t ready to firmly plant my flags on the topic. However, I’ve decided I will repost it in the near future (with some edits, of course, given how fast AI is evolving), but before that, I wanted to tackle a specific aspect of AI that has been front and center in my role: AI and academic integrity.
I’m no expert in AI, though I’m doing everything I can to learn about generative AI and its impact on education. As the individual responsible for overseeing the academic portion of our institution’s Honor Code, I’ve seen firsthand the influence of AI on academic integrity over the last two years. For our institution, we saw a 700% increase in AI-related academic dishonesty cases during the 2022-2023 academic year. This rise, from 2 cases in 2021-2022 to 16 in 2022-2023, was paired with a 55% drop in reported plagiarism cases (34 to 15). This could suggest that students are shifting away from traditional plagiarism and turning toward AI-generated content to complete assignments or instructors may be more focused on the possible use of AI and not as focused on plagiarism (Wily, 2024; NeJame et al., 2023). Regardless, these numbers likely underrepresent AI’s actual use, as proving AI-generated work is difficult and often hinges on student admissions of responsibility rather than concrete evidence (Wily, 2024).
Spoiler alert: I currently have an open mind when it comes to AI. I believe every institution must have active discussions about its ethical use and decide how to educate students on AI. I agree that there are valid and serious concerns when it comes to academic integrity, and what follows is my attempt to share my experience and perspective on just a small portion of this evolving issue.
AI’s Disruption of Academic Integrity
Generative AI has created a seismic shift in how we think about academic integrity. Similar to the initial reactions when institutions introduced online learning, Wikipedia, and broad access to the internet, faculty today are divided over the use of AI. When learning management systems (LMS) began to integrate APIs and automation, similar concerns emerged about reducing student effort and increasing cheating. However, as with these past innovations, we must focus on how to guide students in using AI appropriately, not in banning the tool altogether. Much like Wikipedia, which is now widely accepted as an introductory research tool despite initial skepticism, AI is likely to remain a fixture in education (Mowreader, 2023).
Recent research shows that AI tools like ChatGPT have blurred the lines of academic misconduct, causing institutions to rethink what constitutes cheating. Educators report a significant increase in academic dishonesty cases involving AI. Some studies show that 96% of instructors believe cheating has risen in the past year, compared to 72% in 2021 (Vanderbeek, 2024; White, 2023). Furthermore, the Wily Academic Integrity Report indicates that 45% of students are using AI to assist with writing assignments, while 55% of instructors remain reluctant to incorporate AI into their teaching practices (Wily, 2024).
Interestingly, this reluctance may stem from the challenges of balancing AI’s potential for enhancing learning with the risks to academic integrity. Faculty members are divided on whether AI represents a powerful learning tool or an existential threat to traditional education models. Some educators see AI as an inevitable part of the modern classroom, while others feel unprepared to manage its ethical implications. Faculty polarization is evident, with AI being seen either as a game-changer in pedagogy or a dangerous shortcut that encourages academic dishonesty (D’Agostino, 2023).
The challenge of detecting AI-generated content has made it difficult to fully understand the scope of its use. Several studies have highlighted the limitations of current AI detection tools. Reports show that AI-generated content can often evade detection through simple paraphrasing, reducing accuracy rates to as low as 33% (Sadasivan et al., 2024; Davalos & Yin, 2024). Additionally, 51% of students surveyed in another study said they would continue to use AI tools, even if explicitly prohibited (NeJame et al., 2023). These findings mirror the challenges we face at my institution, where we often rely on student admissions of responsibility once confronted, as definitive proof is hard to achieve with a “more likely than not” standard.
This uncertainty parallels the resistance faced during the early introduction of disruptive technologies like Wikipedia or API integrations in education. Back then, there were concerns about the erosion of traditional academic values. However, over time, we learned to incorporate these tools with proper guidelines. AI today is no different. It requires clear frameworks for ethical usage and thoughtful integration into teaching and learning processes, ensuring students understand its responsible use rather than viewing it as a shortcut.
Adding to this ongoing conversation, President Biden’s Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence, issued on October 30, 2023, underscores the federal government’s recognition of AI’s profound impact. It outlines eight guiding principles, including promoting safety, equity, privacy, and responsible innovation in AI development (U.S. Department of Education, 2024). This executive action encourages institutions to prepare policies that embrace AI’s potential while ensuring its use aligns with ethical standards. As part of the Department of Education’s response, educational institutions are urged to provide clear guidelines for AI use and foster AI literacy among faculty and students alike. Such actions will help us weather the current ‘AI storm’ and develop a more balanced understanding of its role in education, while also preparing modern learners to effectively use AI in their future careers and everyday life.
The Biases and Limitations of AI Detectors
One of the most troubling aspects of AI detection tools is their inherent bias against non-native English writers. Studies reveal that 61.3% of essays written by non-native English speakers were misclassified as AI-generated, while native English essays were accurately classified in most cases (Liang et al., 2023). This bias has significant implications for academic fairness. Non-native speakers are often disadvantaged by these tools, which may inaccurately assess their work based on linguistic patterns rather than genuine misconduct (Wily, 2024). Additionally, AI detectors may flag formulaic writing styles used by students with specific learning disabilities, as demonstrated in Bloomberg’s coverage of a student falsely accused of cheating because of her structured writing and communicative style (Davalos & Yin, 2024).
There is also bias in how faculty identify possible AI use. Often, instructors rely on a hunch or the feeling that “this isn’t how the student writes,” which can introduce unintended bias against students for a variety of reasons. Whether it’s linguistic differences, learning styles, or preconceptions about the student, this kind of subjective judgment can further complicate the fair assessment of academic integrity (Liang et al., 2023; Spencer, 2024). Additionally, AI detection tools can be easily circumvented through paraphrasing or rewording AI-generated text, reducing their effectiveness in identifying AI content (Sadasivan et al., 2024). This is why it’s essential for faculty to engage in open conversations about AI use in their classrooms. Establishing clear guidelines around acceptable AI use can mitigate these biases and reduce unfair assessments (Mowreader, 2023).
The AI-U/1.0: A Student Guide to Navigating College in the Artificial Intelligence Era provides practical advice for students, warning them about the pitfalls of AI tools and emphasizing the importance of original work and citation. The guide also acknowledges that AI detection tools are not infallible and advises students to be transparent about their AI use when required by faculty (Ajjan et al., 2024). These resources are critical for establishing transparent policies and ensuring fairness, particularly for non-native speakers and students with neurodivergent conditions.
Rethinking Teaching, Learning, and Assessments
The rise of AI is pushing educators to reconsider traditional methods of assessment. Written essays, long the gold standard for evaluating critical thinking and communication skills, are now susceptible to AI manipulation. Research shows that 27% of students are using AI tools to help generate written content, leading to concerns about the erosion of critical thinking skills (NeJame et al., 2023). However, the actual use of AI is likely underreported by students, suggesting that the impact of AI on academic integrity may be more widespread than faculty realize (Wily, 2024).
This trend has sparked a movement toward alternative assessment methods that prioritize originality and deeper engagement with the material. Faculty members are experimenting with oral exams, project-based assessments, and in-person evaluations to discourage AI misuse and foster authentic learning (White, 2023; Colby, 2024; Chami, 2023). As AI technologies continue to evolve, the challenge of detecting AI-generated text becomes increasingly complex. A study by Sadasivan et al. (2024) highlights the limitations of current AI detectors, revealing that even sophisticated detection systems can be tricked by advanced paraphrasing tools, which complicates the task of identifying AI-generated text reliably (Sadasivan et al., 2024). This underscores the urgency of rethinking assessments to focus on student originality and active engagement rather than simply relying on traditional written outputs.
Despite these challenges, AI does not necessarily mark the end of written assessments. Rather, it signals the need for innovative approaches. For example, educators can reframe essay assignments to require deeper levels of engagement and personalization. Moving beyond generic or broad prompts, assignments can be more dynamic, integrating multiple stages like brainstorming, outlining, and drafting that encourage students to develop their ideas progressively. This structure not only reduces the temptation to misuse AI tools but also fosters a more authentic writing process, where students can learn to use AI for support in early stages without compromising the integrity of their final work.
Further, the use of AI in assignments can be normalized by explicitly integrating AI as part of the learning process. The AI-U/1.0 Student Guide encourages students to use AI for specific tasks like brainstorming and refining their ideas but cautions against relying on AI-generated content for final submissions. This approach promotes transparency and teaches students responsible AI usage, allowing them to engage with AI critically while ensuring academic integrity (Ajjan et al., 2024).
Recent reports also suggest that students view AI as a valuable tool for learning, not just for completing assignments. Wily (2024) found that 53% of students believe AI helps them understand complex topics more easily, while 36% fear being accused of cheating if they use AI for legitimate academic purposes. Such concerns highlight the need for clear guidelines and policies that help students navigate the ethical use of AI while fostering trust between students and instructors (Wily, 2024).
The ethical use of AI should also be accompanied by institutional efforts to redefine assessments. Some institutions are already experimenting with alternative assessment models, such as in-person oral exams, group projects, and real-world applications that demand critical thinking, collaboration, and creativity. These assessments are designed to make AI misuse difficult while enhancing the learning experience (Mowreader, 2023). Oral exams, for instance, offer a dynamic way to assess student understanding and originality, as they require real-time responses that AI tools cannot easily replicate.
Incorporating AI into the academic process while maintaining integrity also requires educating students on how AI works and the implications of its misuse. This can include transparency around how AI detectors function and their limitations. As Sadasivan et al. (2024) demonstrate, current AI detectors can be easily circumvented by advanced paraphrasing, making it difficult for educators to rely solely on these tools. Instead, institutions must focus on teaching students about responsible AI use and ensuring that assessments are designed to reflect genuine learning outcomes.
In conclusion, the rise of AI challenges traditional teaching, learning, and assessment paradigms. However, with thoughtful integration of AI into the learning process, alongside clear guidelines and innovative assessment strategies, educators can continue to promote academic integrity while embracing the benefits that AI offers.
Embracing AI While Addressing Academic Integrity
While AI poses challenges to academic integrity, it also offers profound opportunities to enhance learning and create more inclusive educational experiences. AI can help personalize education by tailoring learning experiences to students’ unique needs, leading to improved outcomes and more individualized instruction (Chami, 2023). In fact, 52% of students believe AI can effectively help them understand complex topics, especially when integrated into classroom instruction (Colby, 2024).
AI is already making strides in creating accessible learning environments, particularly for students with disabilities. By offering adaptive tools and feedback mechanisms, AI has the potential to fill gaps in traditional teaching, ensuring that all students, regardless of their needs, receive the support necessary for success (Spears, 2023; Ajjan et al., 2024). These tools, such as speech-to-text or adaptive tutoring systems, can provide personalized learning experiences that instructors may not be able to deliver on their own.
In addition to assisting students with disabilities, AI offers the opportunity to reshape how plagiarism and academic standards are viewed. Instead of reinforcing traditional academic boundaries, AI can be used to redefine what is considered creative and innovative work. By incorporating AI literacy into curricula, institutions can better prepare students for the AI-driven workforce, ensuring responsible and ethical use of these tools (Mitrano, 2024).
Institutions like Elon University have already embraced this concept through the AI-U/1.0 guide, which equips students with the knowledge and skills necessary to use AI effectively and ethically in both academic and professional settings. By treating AI as an educational tool rather than a threat, students can develop critical thinking, ethical decision-making, and technical skills that will serve them in their future careers (Ajjan et al., 2024; Chami, 2023).
Faculty, too, are seeing the potential of AI in education. By embracing AI’s role in enhancing learning, faculty can facilitate an open dialogue with students about its ethical use. This inclusive approach can help bridge the divide between those who see AI as a threat and those who view it as an essential learning tool (D’Agostino, 2023).
The Need for Institutional Policy and Faculty Development
It is clear that formalized institutional policies regarding AI use are urgently needed. Current data shows that only 3% of institutions have comprehensive AI policies in place, while 58% are in the process of developing them (NeJame et al., 2023). Without clear guidelines, students and faculty are left to navigate AI’s complexities without adequate support. As the Wily Report suggests, the key to maintaining academic integrity lies in providing clear expectations and fostering a culture of responsible AI use (Wily, 2024).
However, with the rapid integration of AI into everyday tools like Google Workspace and Microsoft Suite, traditional policy frameworks may already be outdated. Rather than focusing solely on creating rigid AI-specific policies, institutions might benefit from developing a broader, more flexible framework. This approach acknowledges that AI is now embedded in the tools we use daily, making it difficult to draw strict boundaries around its use (Justus & Janos, 2024).
For example, many institutions still treat AI as an external tool—something students consciously choose to use, such as ChatGPT or other standalone AI platforms. But as AI becomes seamlessly integrated into word processors, design software, and LMS platforms, it becomes harder to determine when a student is intentionally using AI or merely interacting with technology that incorporates it by default. In light of this, institutions should rethink their approach to AI policies, shifting toward adaptable frameworks that evolve with the technology. This would allow educators and students to work within an AI-driven environment while maintaining academic integrity (Justus & Janos, 2024).
Spencer and Ajjan et al. (2024) both highlight the importance of institutions taking proactive steps to train faculty on AI tools and their limitations. Faculty development is essential to ensure that educators are well-equipped to redesign their assessments and incorporate AI literacy into their teaching (Ajjan et al., 2024; White, 2023). The EDUCAUSE Review emphasizes that professional development programs must help faculty reimagine their courses to mitigate AI misuse while promoting its positive applications (White, 2023).
Furthermore, clear expectations for AI use must be established to prevent ambiguity. Faculty should be equipped not only to understand AI’s capabilities but also to craft assignments that allow students to use AI ethically while ensuring that their own academic work remains authentic (Mowreader, 2023). Providing educators with access to AI tools and professional development opportunities ensures that they are prepared to handle this evolving landscape.
By adopting a framework-based approach rather than relying solely on static policies, institutions can create an environment that supports responsible AI use and helps prepare students for a world where AI will be ubiquitous. This strategy allows for ongoing adaptation, ensuring that institutions remain agile in the face of rapid technological change (Justus & Janos, 2024).
Final Thoughts
As we move forward in this AI-driven world, it’s clear that AI is here to stay, and its role in education will only continue to grow. The key is not to fear it but to find ways to work with it—helping our students, educators, and institutions benefit from AI’s capabilities while protecting the core values of academic integrity.
AI offers incredible opportunities to personalize learning, assist students with unique needs, and open up new ways of teaching and learning. If we incorporate AI literacy into our courses, we’re not just preparing students for the future; we’re teaching them how to use these tools responsibly and ethically. AI can be a powerful partner in education, but only if we approach it thoughtfully, with clear boundaries and a strong ethical foundation.
At the same time, we can’t ignore the challenges. AI-generated content has blurred the lines of what’s considered academic dishonesty, and traditional assessments like essays or take-home exams are now vulnerable to misuse. The solution isn’t simply better detection tools—AI will keep evolving. Instead, we need to rethink how we assess students, focusing on assignments that encourage originality and deeper engagement, while being transparent about how AI can and can’t be used.
Faculty play a crucial role in this shift. With the right training and resources, educators can redesign their teaching to integrate AI in a way that supports learning without compromising integrity. This isn’t just about changing individual courses—it’s about building a culture of responsible AI use across institutions.
And institutions need to step up, too. The fact that only 3% of schools have clear AI policies in place (NeJame et al., 2023) highlights the urgency for developing guidelines that make sense in today’s world. Schools need to set clear expectations for both students and faculty, offering support and consequences where necessary, so that AI is seen as a tool for innovation, not a way to cut corners.
Ultimately, the future of AI in education will depend on how well we adapt. AI isn’t a threat to learning—it’s an opportunity to teach our students how to think critically, engage with technology responsibly, and prepare for the evolving workforce. But to do this right, we need to make sure that AI is used in ways that uphold our core values—integrity, fairness, and inclusion.
As we embrace AI, we must remain committed to these principles. By working together—through policies, faculty development, and ongoing conversations with students—we can make sure that AI enhances, rather than undermines, the integrity and excellence of higher education.
References
Ajjan, H., Akben, M., Alexander, B., Anderson, D. J., Book, C., et al. (2024). AI-U/1.0: A student guide to navigating college in the artificial intelligence era. Elon University. https://www.elon.edu/u/news/2024/08/19/student-guide-to-ai/
Chami, G. (2023, October 23). Artificial intelligence and academic integrity: Striking a balance. THE Campus. Retrieved from https://www.timeshighereducation.com
Colby, E. (2024). AI has hurt academic integrity in college courses but can also enhance learning. Wiley. https://newsroom.wiley.com/press-releases/press-release-details/2024/AI-Has-Hurt-Academic-Integrity-in-College-Courses-but-Can-Also-Enhance-Learning-Say-Instructors-Students/default.aspx
D’Agostino, S. (2023, September 13). Why professors are polarized on AI. Inside Higher Ed. Retrieved from https://www.insidehighered.com/news/tech-innovation/artificial-intelligence/2023/09/13/why-faculty-members-are-polarized-ai
Davalos, J., & Yin, L. (2024, October 18). Do AI detectors work? Students face false cheating accusations. Bloomberg Businessweek. Retrieved from https://www.bloomberg.com/news/features/2024-10-18/do-ai-detectors-work-students-face-false-cheating-accusations
Justus, Z., & Janos, N. (2024, October 22). Your AI policy is already obsolete. Inside Higher Ed. Retrieved from https://www.insidehighered.com/opinion/views/2024/10/22/your-ai-policy-already-obsolete-opinion
Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., & Zou, J. (2023). GPT detectors are biased against non-native English writers. Pattern, 4(7), 1-4. https://doi.org/10.1016/j.patter.2023.100779
Mitrano, T. (2024, January 16). Plagiarism, AI and higher education. Inside Higher Ed. Retrieved from https://www.insidehighered.com/opinion/blogs/law-policy-and-it/2024/01/16/plagiarism-ai-and-higher-education
Mowreader, A. (2023, September 28). Academic success tip: Establish guidelines for AI use. Inside Higher Ed. Retrieved from https://www.insidehighered.com/news/student-success/academic-life/2023/09/28/report-three-ways-address-generative-ai-college
NeJame, L., Bharadwaj, R., Shaw, C., & Fox, K. (2023, April 25). Generative AI in higher education: From fear to experimentation, embracing AI’s potential. Tyton Partners. Retrieved from https://tytonpartners.com/generative-ai-in-higher-education-from-fear-to-experimentation-embracing-ais-potential/
Sadasivan, V., Kumar, A., Balasubramanian, S., Wang, W., & Feizi, S. (2024). Can AI-generated text be reliably detected? University of Maryland. Retrieved from https://arxiv.org/pdf/2303.11156
Spears, A. (2023). AI as an ally: Enhancing education while upholding integrity. Retrieved from https://www.literacyworldwide.org/blog/literacy-now/2024/10/01/ai-as-an-ally-enhancing-education-while-upholding-integrity
U.S. Department of Education, Office of Educational Technology. (2024). Designing for education with artificial intelligence: An essential guide for developers. U.S. Department of Education. https://tech.ed.gov/files/2024/07/Designing-for-Education-with-Artificial-Intelligence-An-Essential-Guide-for-Developers.pdf
White, J. (2023, November 6). Academic integrity in the age of AI. EDUCAUSE Review. Retrieved from https://er.educause.edu/articles/sponsored/2023/11/academic-integrity-in-the-age-of-ai
Wily. (2024). The Wily academic integrity report: Instructor and student experiences, attitudes, and the impact of AI. https://www.wiley.com/en-us/network/education/instructors/teaching-strategies/the-latest-insights-into-academic-integrity-instructor-and-student-experiences-attitudes-and-the-impact-of-ai-2024-update