The cyclical ethical effects of using artificial intelligence in education

Area of Science:

• Educational Technology
• Artificial Intelligence Ethics
• Sociology of Education

Background:

• The integration of Artificial Intelligence (AI) in education presents complex ethical challenges.
• Existing literature highlights several key areas of concern regarding AI's impact on learning environments.

Purpose of the Study:

• To conduct a synthetic review of literature on the ethics and effects of AI in education.
• To identify and analyze five distinct, interrelated divides stemming from AI implementation in educational settings.

Main Methods:

• A comprehensive review and synthesis of relevant academic literature.
• Qualitative analysis to identify distinct themes and interrelationships between identified divides.

Main Results:

• Identified five key divides: access, representation, algorithms, interpretation, and citizenship.
• Analyzed upstream divides (access, representation) and downstream divides (interpretation, citizenship) in relation to algorithmic bias.
• Highlighted how these divides perpetuate structural biases in teaching and learning.

Conclusions:

• Without proactive human intervention, AI in education can create a vicious cycle of bias.
• Mitigating bias through increased human responsibility and control can foster a virtuous cycle for diversity, equity, and inclusion.
• Recommendations for progressive improvement to enhance educational opportunity and effectiveness for all learners.