While academic misconduct, particularly plagiarism, has long been a concern for educators, the methods students use have evolved significantly. Recent trends are intensifying the risk to institutional reputation, posing critical questions about how institutions define and safeguard originality. The rapid integration of AI is no longer just about text; it's about AI-generated code, sophisticated methods designed to bypass detection, and new forms of contract cheating.
This evolution is reshaping how educators teach and design assessments while encouraging students to uphold academic integrity.
The data quantifies this evolution. 2025 research from HEPI reveals that 18% of UK undergraduate students admit to submitting AI-generated text in their assignments, while a broader 2025 study by Turnitin and Vanson Bourne found that 95% of the academic community believes AI is being misused at their institutions, illustrating a significant shift in both the methods and awareness of academic misconduct.
Our goal is to provide insights into the evolving risks and help you protect the integrity of your institution.
How are new and emerging trends reshaping academic misconduct?
Recent trends are intensifying the risk to institutional reputation, posing critical questions about how institutions define and safeguard originality in student work. The rapid integration of AI tools is reshaping how educators conduct their teaching and design assessments while encouraging their students to uphold the fundamental values of academic integrity.
As we explore these evolving challenges, we will examine various emerging forms of misconduct, from generative AI misuse and automated text modification to contract cheating and sophisticated exam tactics. Understanding these trends is crucial for adapting our approaches to maintain academic standards and integrity in this dynamic environment.
Generative AI misuse
The advent of generative AI has introduced new dimensions to education, particularly, academic misconduct. As institutions embrace AI for enhancing teaching and learning, they also face challenges.
The challenge to originality and detection
AI tools, designed to generate and refine text, can be used for legitimate learning, including research and idea generation, as well as learning and tutoring. However generative AI can be misused by students to produce and submit work that appears original but is actually created by algorithms.
The sophistication of generative AI tools makes it difficult for assessors to distinguish between human-written and AI-generated content, complicating the task of maintaining academic standards - particularly without the support of AI writing detection technology.
The risk of fabricated sources
Generative AI tools, such as ChatGPT, can further complicate proper citation by providing students with fabricated or inaccurate sources. While these tools can generate text and suggest sources, they may also invent references or create plausible-sounding but non-existent sources (also coined Ai hallucinations), leading to a new form of source-based plagiarism.
As AI-generated content becomes more sophisticated, the risk of students inadvertently or deliberately using fabricated sources increases, making it crucial for educators to stay vigilant and implement robust citation practices.
The impact on critical thinking
As students become more reliant on AI-generated content, there is concern that this may hinder the development of essential academic skills. This concern is shared by students themselves; a 2025 study by Turnitin and Vanson Bourne found that 59% of students worry that over-reliance on AI could lead to a reduction in critical thinking skills.
Writing for Forbes, leadership expert, Ron Carucci says, “No clever prompt we type into an AI tool will ever be able to replace human critical thinking … In reality, critical thinking becomes even more necessary in the age of AI, both to use it properly, and to do the necessary work behind the scenes to make it a more reliable tool.”
The global policy dilemma
The ethical and cultural dynamics surrounding the use of generative AI differ globally. Research by Jin et al. (2024) highlights geographic differences in AI adoption. Their study of 40 universities found that while five institutions in the UK and Australia focus on maintaining academic integrity and originality, eight universities in the US and Hong Kong emphasize leveraging AI to enhance teaching and learning.
This disparity poses a challenge for developing universal standards and policies. Institutions must navigate these varied approaches to create strategies that not only address the misuse of generative AI but also harness its potential to improve learning, ensuring a balanced approach that upholds academic standards worldwide.
Automated text modification
Automated text modification involves using software to alter existing content to evade plagiarism detection systems. One basic method is text spinning, where words or phrases are replaced with synonyms, and sentence structures are adjusted using simple algorithms. This often results in text that remains somewhat recognizable as derived from the original source.
A more sophisticated form of automated text modification is AI paraphrasing. AI paraphrasing tools employ advanced algorithms to understand and rephrase text while preserving its original meaning, creating content that appears more original and coherent. While serving different purposes, both text spinning and AI paraphrasing challenge traditional methods of assessing the authenticity of academic work, making it more and more difficult for instructors to identify unoriginal submissions.
Beyond AI paraphrasing, a more deceptive trend for 2025 is the use of AI 'humanizers' or 'bypasser' tools. These services are specifically designed to rewrite AI-generated text to evade detection. They do this by intentionally varying sentence structures, introducing stylistic imperfections, and altering word choices to mimic human writing patterns more closely.
This represents a direct cat-and-mouse game, where the goal isn't just to rephrase content but to actively conceal the involvement of AI, posing a significant and growing threat to academic integrity.
Sophisticated code plagiarism
In technical fields, AI coding assistants like GitHub Copilot and other large language models are reshaping how software is developed. While these tools are invaluable for professional developers, they also introduce a new vector for academic misconduct. Students can now generate complex code, entire functions, or even complete programming projects with minimal effort and understanding.
This sophisticated code plagiarism is difficult to detect because the AI-generated code is often functional and may not exist in any previously submitted work, making it a distinct challenge from traditional code copying. It questions not just the originality of the submission but whether the student has genuinely acquired the core programming skills being assessed.
Contract cheating
Also known as ghostwriting, contract cheating generally involves students hiring third parties to complete their assignments or exams on their behalf. This practice is facilitated by the widespread availability of affordable online services that offer quick assistance. Despite some countries implementing laws to combat contract cheating by banning the advertisement of essay mills, illegal cheating websites and social media accounts persistently find their way to students.
Since the Australian government introduced its essay mill law in 2021, the Tertiary Education Quality and Standards Agency (TEQSA) reports that over 300 illegal cheating websites have been blocked, as of May 2025. While contract cheating might appear to be an “old school” form of academic misconduct, it remains a significant concern.
The emergence of generative AI has only further refined and expanded essay mill services, making them more sophisticated and even harder to detect. The challenge of contract cheating is also evolving from a battle against known essay mills to a broader issue of verifying authorship. With the rise of AI, a ghostwriter can now produce text that is stylistically neutral and harder to distinguish from a student's own work. In high-stakes cases, this raises critical questions about who truly authored a submission, pushing the boundaries of traditional detection methods and increasing the need for tools and strategies that can help confirm authorship.
Contract cheating can also manifest itself in the form of impersonation, where a student hires someone else to take a test or even their entire university course. They may even use third-party tutors to provide answers during a live (usually online) exam.
What are the new and emerging trends in exam misconduct?
In addition to plagiarism, exam misconduct has evolved with the adoption of new technologies and methods are adopted by students seeking unfair advantages in high-stakes settings. Traditional forms of exam cheating, such as using unauthorized notes or copying from peers, are now being complemented by more sophisticated tactics.
A notable trend is the use of digital devices and apps to gain access to unauthorized information. For instance, students might use hidden earpieces or smartwatches to receive answers during exams. The proliferation of online forums and social media platforms has also enabled the sharing of exam questions and answers, further complicating efforts to maintain exam integrity.
Another emerging trend is the use of generative AI tools to assist in real-time exam cheating. Students may use AI to quickly generate responses, exploiting the technology’s ability to produce coherent and contextually relevant text. This has led to increased scrutiny on the use of AI tools and a call for updated exam policies that address these new challenges. A study conducted by Scarfe et al. (2024) at the University of Reading revealed that ChatGPT-generated exam answers went undetected in 94% of cases, achieving higher grades than actual student submissions on average. This highlights the growing challenge of detecting AI-assisted exam misconduct.
To counter these trends, institutions are adopting advanced proctoring technologies and updating their academic integrity policies. Enhanced monitoring during exams, such as secure exam software that offers AI-driven surveillance, are becoming more common in an effort to detect and prevent misconduct.
Which more traditional forms of academic misconduct should educators continue to look out for?
While new and emerging trends in academic misconduct are important to address, it is crucial for educators to also remain vigilant about traditional forms of misconduct that have persisted over time. Among these, word-for-word plagiarism stands out as the most basic and widely recognized form. This involves directly copying text from a source without proper citation and remains a fundamental concern in academic integrity.
However, the landscape of academic misconduct is diverse, and several traditional forms continue to pose significant challenges…
Self-plagiarism
Self-plagiarism—sometimes known as “duplicate plagiarism”—occurs when a student recycles their own previously submitted work without proper acknowledgment. While it may seem harmless, it is considered academic misconduct because it involves presenting old work as new, which can mislead educators about a student's current efforts.
AI-assisted plagiarism (patchwriting 2.0)
A hybrid form of misconduct is emerging where students use generative AI as a sophisticated paraphrasing or research tool but fail to properly engage with or cite the sources. In this scenario, a student might ask an AI to "explain a concept using these three articles," then copy the AI-generated summary into their paper. The result is a complex mosaic of AI-generated text and improperly attributed ideas from existing sources.
Student collusion
Collusion involves students working together inappropriately on assignments meant to be completed individually. In today’s learning environments, digital communication tools can make collusion easier to facilitate, as students can quickly share answers or divide tasks via messaging apps or shared documents.
Data plagiarism
Considered one of the most severe forms of academic misconduct, data plagiarism undermines the integrity of academic findings and the trust placed in them. It can take the form of copying data sets from other sources without proper attribution, fabricating data, or manipulating existing data to fit desired outcomes. By plagiarizing data, the act becomes a form of fabrication, as the data is only ever valid in its original context (Dougherty, 2020).
In an analysis of over 7,500 articles, Phogat et al. (2023) found that data fabrication was most prevalent in nonself-reported studies, indicating that much of this type of misconduct may go unreported or undisclosed by those involved. This highlights the challenge of detecting and addressing it.
For students, data plagiarism often manifests in assignments that require original research or analysis, whereby students might copy data from published papers, websites or other students’ work, presenting it as their own. However, it is in research that the consequences are far wider in reach. When researchers plagiarize or falsify data, it creates a ripple effect of misinformation, which can have catastrophic outcomes, particularly in fields like medical research.
From detection to clarity in the face of new and emerging trends in academic misconduct
Understanding and addressing new trends in academic misconduct is crucial, and tools that help ensure the authenticity of student work remain a vital part of upholding academic standards.
However, the most resilient integrity strategy combines effective technology with proactive pedagogy.
The focus is expanding from solely detecting misconduct to actively guiding students toward ethical and effective AI use. This is where a 'show your work' approach becomes invaluable. By designing assessments that require students to share process artifacts—like drafts, outlines, and reflections—educators gain a transparent view into how a student’s work was constructed.
This approach is not about deterring AI use, but preventing AI misuse. It shifts the conversation from a punitive "Did AI write this?" to a developmental "How did you use AI to support your learning?"
By making the writing process visible, educators can reinforce authentic skill development while teaching the AI literacy students need for their future careers. Prioritizing this transparency builds a durable culture of academic integrity prepared for the challenges ahead.
This article was originally published on September 19, 2024, and was last updated on October 14, 2025, to include the latest trends and research in academic misconduct.