The rapid development of technology, particularly artificial intelligence (AI), has brought many benefits to our daily lives. However, this technological advance is not without consequences for academic integrity. In the age of AI, the opportunities for cheating have multiplied, challenging educational institutions.
Contents
Cheating has always existed in educational environments, but AI has added a new dimension to the problem. "Traditional cheating" often involves copying work without crediting the author - plagiarism.
AI now offers new possibilities. Students can use AI-based tools to generate content. Students believe that the answers provided by text-generating AI are original and free from plagiarism. And that's the problem.
ChatGPT does not cite its sources most of the time.
It rephrases, uses quotations and copies the authors' writing style without crediting them. This is considered plagiarism. This is the problem for authors, but also for researchers, teachers and students.
"GPT templates can unintentionally facilitate plagiarism by generating content that closely resembles existing works without proper attribution. This can lead to ethical and legal problems, particularly in academic, journalistic or creative contexts. (...) whether the intellectual property rights belong to the user, the creators of the model or a combination of both.
Finally, in some cases, GPT models can be trained on proprietary or confidential data, which can lead to the unauthorised use or disclosure of sensitive information.", "The Dark Side of the Moon" by ChatGPT, Alain Goudey, 16/03/2023.
Several factors may be contributing to the rise in cheating with ChatGPT. Firstly, increasing academic pressure is prompting some students to look for quick ways to get good grades, whatever the consequences.
In addition, the ease of access to text-generating AI encourages students to use it without really understanding how to do so with integrity.
Finally, the feeling that "no one will be able to tell that the text was written by ChatGPT" reduces the perception of the risks and penalties associated with cheating.
How can I check whether a text comes from an AI? Is ChatGPT detectable? Despite the challenges, several solutions can be implemented to preserve academic integrity in the age of AI.
First of all, it is essential to make students aware of the consequences of cheating and to promote a culture of intellectual honesty. Academic integrity education programmes can help students understand the fundamental values of ethics and integrity.
Secondly, the pedagogical approach can be adapted to encourage authentic learning and creativity. By focusing on tasks that involve critical thinking, the application of knowledge and problem solving, teachers can reduce the incentive to cheat.
Finally, plagiarism detection tools need to be improved to identify advanced forms of AI-based cheating. Content generated by ChatGPT is detectable. Institutions need to invest in sophisticated detection systems that can spot suspicious similarities and indicators of cheating.
Compilatio already offers this service.
How to cite the use of ChatGPT?
When using ChatGPT in an academic assignment, it is important to cite it to avoid being accused of plagiarism.
To do this, you need to follow the appropriate citation standards for your field of study or your institution's specific guidelines. If there are no standards, you can choose how you cite your source. It must be identical for all referenced sources.
OpenAI. (Year). ChatGPT [Computer software]. Retrieved from https://www.openai.com/research/chatgpt
OpenAI. "ChatGPT." Computer software. OpenAI, year. Web. [Retrieval date]. https://www.openai.com/research/chatgpt
It is important to mention the sources used in your academic work to credit the creators of ChatGPT.
Cheating in the AI age poses unique detection challenges. AI systems can generate sophisticated content that can be difficult to distinguish from genuine, personal work.
In addition, cut-and-paste content from ChatGPT may contain plagiarism and use unauthorised content. Plagiarism detection software makes it possible to find the sources associated with quotations.
Finally, the algorithms used by plagiarism detection tools are not always adapted to detect advanced forms of cheating based on AI. Teachers and institutions therefore face a constant challenge to update their detection methods and keep up with the latest technological advances.
In the age of AI, cheating represents a major challenge to academic integrity. However, by raising student awareness, improving detection tools and promoting a pedagogical approach focused on authentic learning, we can preserve academic integrity. Collaboration between students, teachers and institutions is essential to meet this challenge and ensure that AI is used ethically and responsibly in our educational environments.
To find out more:
- "Should ChatGPT Be Banned From Classrooms?", Bill Of Rights Institute, consulted on 01/08/2023.
- "Exploring new opportunities with AI in education", Microsoft, consulted on 01/08/2023.
- "We asked ChatGPT: Should Schools Ban You?", Education Week, consulted on 01/08/2023.
Information: this informative article, which does not require any personal reflection, was written in part with the help of ChatGPT. We have reworked the automatically-generated content (correcting repetitions, correcting turns of phrase, adding clarifications, adding quotations, checking the veracity of the information, etc.).
