Education and Training Assessment and Artificial Intelligence. A Pragmatic Guide for Educators

Area of Science:

• Higher Education
• Artificial Intelligence
• Educational Assessment

Background:

• Emerging Generative AI tools, including ChatGPT, pose challenges to the validity of current assessment methods in higher education.
• Concerns exist regarding students' potential misuse of AI tools to complete academic assignments.

Purpose of the Study:

• To review existing evidence on the impact of Generative AI on higher education assessments.
• To explore alternative assessment strategies and the importance of evaluating core knowledge.
• To consider these issues within the context of professional regulations for Biomedical Scientists.

Main Methods:

• Literature review of current evidence on Generative AI and academic assessments.
• Analysis of AI capabilities in completing common assessment types (essays, multiple-choice exams).
• Consideration of professional regulatory frameworks (e.g., Health and Care Professions Council).

Main Results:

• Evidence confirms that Generative AI tools like ChatGPT can successfully complete essays and multiple-choice exams to a high standard.
• These AI tools perform assessments quickly and cost-effectively.
• The validity of traditional assessment methods is legitimately challenged by AI capabilities.

Conclusions:

• The findings highlight the need for higher education institutions to adapt assessment strategies in response to AI.
• Alternative assessment methods must be developed to accurately measure student learning.
• Maintaining assessments of foundational knowledge remains crucial for professional competence.