The Stack Overflow Podcast

Is AI making your code worse?

Episode Summary

Ben and Ryan are joined by Bill Harding, CEO of GitClear, for a discussion of AI-generated code quality and its impact on productivity. GitClear’s research has highlighted the fact that while AI can suggest valid code, it can’t necessarily reuse and modify existing code—a recipe for long-term challenges in maintainability and test coverage if devs are too dependent on AI code-gen tools.

Episode Notes

GitClear is a developer-friendly code review tool that aims to deliver higher developer satisfaction and faster releases. Check out their blog or find them on GitHub.

GitClear’s research focuses on how AI code-gen tools have impacted code quality (and not in a good way).

Find Bill on LinkedIn.

Chapters

00:00 Introduction 

00:30 Background of the Research

06:09 Business Model of GitClear

09:46 Copy Pasted Code

10:26 Churn Code

12:21 Code Readability

14:12 Code Suggestions and Auto-Completion

16:34 Drop in Moved Code

23:18 Larger Token Windows

26:31 Improving Gen AI

28:46 Conclusion