# Computing Education in the Era of Generative AI

**Authors:** Paul Denny, James Prather, Brett A. Becker, James Finnie-Ansley, Arto Hellas, Juho Leinonen, Andrew Luxton-Reilly, Brent N. Reeves, Eddie Antonio Santos, Sami Sarsa  
**arXiv:** [2306.02608](https://arxiv.org/abs/2306.02608)  
**Date:** 2023-06-05  
**Summary page:** https://www.summarizepaper.com/en/paper/2306.02608v1/

## Key points

- Computing education community has a history of innovative teaching methods
- Recent advancements in AI have introduced code generation models
- Concerns about the impact of these models on society and computing education
- Paper titled "Computing Education in the Era of Generative AI" discusses challenges and opportunities
- Findings from two recent articles evaluating code generation models' performance and learning resource quality
- Analysis of impacts on pedagogical practices, highlighting positive opportunities and potential challenges
- Insights into how code generation models can shape and transform introductory programming classrooms

## Summary

The computing education community has a long history of innovative teaching methods to support students in introductory courses and assist teachers in facilitating student learning. Recent advancements in artificial intelligence (AI) have introduced code generation models that can accurately produce source code from natural language problem descriptions. These models have become widely available and easy to use, raising concerns about their potential impact on various aspects of society, including the future of computing education. In this paper titled "Computing Education in the Era of Generative AI," authors Paul Denny, James Prather, Brett A. Becker, James Finnie-Ansley, Arto Hellas, Juho Leinonen, Andrew Luxton-Reilly, Brent N. Reeves, Eddie Antonio Santos and Sami Sarsa discuss the challenges and opportunities presented by these code generation models specifically in introductory programming classrooms. The authors summarize the findings of two recent articles. The first article evaluates the performance of code generation models on typical introductory-level programming problems and finds impressive accuracy in generating source code from natural language problem descriptions. The second article explores the quality and novelty of learning resources generated by these AI models. Considering the most recent advances at the time of writing, they analyze the likely impacts of these models on pedagogical practices; highlighting both positive opportunities and potential challenges that arise with their integration into computing education. This paper provides valuable insights into how code generation models powered by AI can shape and transform introductory programming classrooms. By examining their performance on programming problems and evaluating the quality of learning resources they generate educators can better understand how to leverage these technologies effectively while addressing any concerns or limitations they may present.