Skip to main content Skip to navigation

The Responsible Development of Automated Student Feedback with Generative AI

Project Overview

The document explores the transformative potential of generative AI, especially large language models (LLMs), in enhancing student feedback within the educational landscape by providing scalable, repeatable, and instantaneous responses. It underscores the critical ethical considerations associated with deploying such technologies, particularly the necessity for inclusivity and equity to ensure all students benefit equally. The importance of human domain expertise is emphasized as essential for developing AI models that remain relevant and effectively cater to the diverse needs of students. Furthermore, the document raises valid concerns regarding the implications of automated feedback on learning processes, advocating for responsible development and ongoing monitoring of AI systems to ensure they positively impact education. Overall, it highlights the dual need for innovation in educational tools through generative AI while simultaneously prioritizing ethical standards and human oversight to foster an equitable learning environment.

Key Applications

Automated Feedback Generation using Generative AI

Context: Higher education in various programs, particularly in STEM fields, where students seek personalized feedback on assignments.

Implementation: Utilizing large language models (LLMs) such as ChatGPT to generate personalized and authentic feedback based on previous student submissions and assignments, leveraging machine learning models that continuously refine their outputs.

Outcomes: ['Potential for abundant, scalable feedback that supports diverse learning needs.', 'Increased efficiency in providing feedback, potentially leading to improved student outcomes.']

Challenges: ['Risk of overlooking unique learner needs.', 'Ethical considerations related to AI feedback.', 'Concerns about factual accuracy and bias in training data.', 'Need to address uncommon student submissions and continuous model refinement.']

Implementation Barriers

Ethical Barrier

Challenges related to ensuring inclusivity and equity in AI-generated feedback, particularly for diverse student populations. This includes addressing ethical considerations in the use of AI in education.

Proposed Solutions: Develop a framework for responsible AI use that addresses ethical considerations and emphasizes inclusivity.

Technical Barrier

Limitations of current LLMs in providing accurate and relevant feedback, especially for unique or uncommon submissions. This includes the need for continuous monitoring and refinement of AI models.

Proposed Solutions: Continuous monitoring and refinement of AI models, along with the development of tailored datasets for training.

Operational Barrier

The need for significant resources and expertise to develop and maintain effective AI feedback systems, combined with the necessity to employ transfer learning methodologies and utilize existing educational materials to reduce costs.

Proposed Solutions: Employ transfer learning methodologies and utilize existing educational materials to reduce costs.

Project Team

Euan D Lindsay

Researcher

Mike Zhang

Researcher

Aditya Johri

Researcher

Johannes Bjerva

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Euan D Lindsay, Mike Zhang, Aditya Johri, Johannes Bjerva

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies