# Exploring the Use of ChatGPT as a Tool for Learning and Assessment in Undergraduate Computer Science Curriculum: Opportunities and Challenges

**Authors:** Basit Qureshi  
**arXiv:** [2304.11214](https://arxiv.org/abs/2304.11214)  
**Date:** 2023-04-16  
**Summary page:** https://www.summarizepaper.com/en/paper/2304.11214v1/

## Key points

- Artificial Intelligence (AI) in teaching and learning has gained significant attention in computing education
- ChatGPT, an AI-based tool, enhances student involvement, cooperation, and accessibility
- The study focuses on using ChatGPT for learning and assessment in undergraduate Computer Science curriculum, specifically for fundamental programming courses
- Two groups of students were given programming challenges, with one group having access to ChatGPT and the other group having no internet access
- Students using ChatGPT had an advantage in terms of earned scores compared to the control group
- Inconsistencies and inaccuracies were observed in the submitted code by students using ChatGPT
- Incorporating AI tools like ChatGPT into higher education presents opportunities and challenges
- There is a need to address issues related to accuracy and consistency when utilizing AI-based tools for learning and assessment purposes
- Integrating AI into undergraduate Computer Science curriculum through tools like ChatGPT can enhance teaching and learning experiences while navigating complexities introduced by AI technologies.

## Summary

The use of Artificial Intelligence (AI) in teaching and learning has gained significant attention in the field of computing education. One AI-based tool, ChatGPT, has shown promise in enhancing student involvement, cooperation, and accessibility. This paper focuses on exploring the potential benefits and challenges of using ChatGPT as a tool for learning and assessment in undergraduate Computer Science curriculum, specifically for fundamental programming courses. The study involved students who had completed a Data Structures and Algorithms course at the sophomore level. Two groups of students were given programming challenges to solve within a limited time frame. The control group (group A) had access to textbooks and notes related to programming courses but no internet access. On the other hand, group B students were provided with access to ChatGPT and encouraged to utilize it for assistance in solving the programming challenges. The challenge took place in a computer lab environment using the PC2 platform. Each team of students approached the problem by writing executable code that satisfied a specific number of test cases. Their performance was evaluated based on the number of successfully passed test cases. The results revealed that students using ChatGPT had an advantage in terms of earned scores compared to the control group. However, there were inconsistencies and inaccuracies observed in the submitted code, which ultimately affected overall performance. After conducting a thorough analysis, this paper concludes that incorporating AI tools like ChatGPT into higher education presents both opportunities and challenges. While it enhances student engagement and accessibility to resources, there is also a need for addressing issues related to accuracy and consistency when utilizing AI-based tools for learning and assessment purposes. Overall, this research sheds light on the potential benefits and obstacles associated with integrating AI into undergraduate Computer Science curriculum through tools like ChatGPT. It provides valuable insights for educators seeking innovative ways to enhance teaching and learning experiences in programming courses while navigating the complexities introduced by AI technologies.