Ways to combat AI-enabled cheating
One of ASQA’s current Regulatory Risk Priorities is to detect academic cheating. Academic cheating has always been a challenge, but the rise of AI tools has added a new layer of complexity. As RTOs, it’s essential to stay ahead of the curve and ensure that our assessments remain robust, authentic, and resistant to misuse.
Here are some strategies to enhance assessment design and tackle AI-enabled cheating:
Focus on Practical and Oral Assessments: Moving beyond written tasks to practical demonstrations and oral questioning can help assess genuine understanding and skills. These methods are harder to falsify and provide valuable insights into a student’s competency.
Personalise Assessments: Tailor tasks to reflect the unique contexts or experiences of each student. By requiring learners to draw on their individual workplace or learning environment, assessments become more authentic and less prone to generic answers.
Leverage Technology Against Cheating: Another option is to utilise AI as an ally to discover AI-enabled cheating. Tools like plagiarism detection software and systems that verify the originality of work may help ensure assessments are genuine, though I would use caution with using such tools. Remember, AI is still a learning process in motion, and makes mistakes.
Encourage Critical Thinking: Design tasks that require analysis, problem-solving, and reflection. These skills are difficult for AI tools to mimic effectively and promote deeper learning for students.
Embed Ongoing Assessment: Continuous assessment throughout the course, rather than relying on a single submission, allows trainers to monitor progress and identify inconsistencies in a student’s performance.
Cheating doesn’t just undermine qualifications; it compromises the safety and trust of workplaces relying on skilled professionals. By prioritising innovative assessment design, we can ensure that our RTOs deliver graduates who are truly job-ready.
How is your RTO redesigning assessments to maintain integrity in the AI era? Share your approaches below.
