Cheating in university assessments has always been a longstanding issue, but it seems to be getting worse with the advancement of technology. The release of free-to-use generative AI tools like ChatGPT has made cheating easier than ever before. Students can plug any assessment questions into the tool and receive a solution, which may or may not be correct.
To understand the impact of ChatGPT on marking processes, a university ran an experiment. Volunteer markers were asked to grade synthetic scripts generated by ChatGPT, alongside real scripts written by students. The markers assessed four end-of-module assessments from various levels of study. The results were eye-opening; almost every undergraduate script, except for two outliers, received at least a passing grade.
To understand the quality and accuracy of ChatGPT generated scripts, the sample scripts were run through diverse quality assurance software, and markers were interviewed. The study proved that ChatGPT is at least capable of producing adequate solutions to a range of questions across a broad level of study.
With the public release of AI tools, they are raising questions of academic integrity and potentially impacting the quality assurance process. The technology is a game-changer that may alter the way marking and grading is done in universities, so it’s essential to have a baseline understanding of its impact.
In conclusion, researchers warn that this technology could potentially undermine the value of degrees in the long run. Therefore, it’s crucial to come up with measures to detect such cheating methods. It is a vital responsibility of all universities and educational institutes to ensure that academic integrity is maintained and upheld in all circumstances.
