Using GenAI to Code? Not So Fast

AI coding agents have been sold to us, making big promises of how much faster we will work using them. In our “The Real Cost of AI-Generated Code: It’s Not All It’s Cracked Up To Be” article, we showed that this speed is coming at a cost with increased Technical Debt and decreased maintainability. The Agile community has witnessed this many times over the years.

Aside from sharpening our skills, most things that promise to help us go faster come at cost. Now there is new research that calls into question even the speed-up of writing code with GenAI.

Read for Yourself and Decide

Summary and the full paper: Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity.

Caveats: this is a single research paper which studied senior developers who contributed to open source projects. The codebases for the projects were large (over 1,000,000 lines of code), and the developers were familiar with their projects. Some effort was made to ensure that the projects had good-quality code in the first place. So the paper is studying a narrow slice of work, and the results might not generalize.

The researchers paid the participants to address outstanding issues in their projects. For each issue, they were told to use AI or no-AI. They were also asked to forecast how long the issue would take to fix. When allowed to use AI, the developers predicted they would go 25% faster than an unaided human. That’s not what happened. The AI-enabled developers took, on average, 19% longer to complete the task. Interestingly, even though they were notably slower than their non-AI aided colleagues, they perceived that they had worked 20% faster.

These results are contrary to what we see reported elsewhere.

More caveats on the research paper:

  • There is a risk that the paper’s methodology missed something important about real work (although this paper already has a better methodology than previous papers).
  • Many of the developers hadn’t used Cursor before. Maybe with more practice they would have been faster.
  • The study used senior developers. Maybe junior developers would gain more benefit from AI assistance. But then that leaves the question of how do junior developers grow and become senior? Will junior developers spend enough time reviewing the generated code, understand it, and verify that it’s a good fit?
  • This is a single paper. Maybe a larger sample size or a diffent mix of people might have found a different result.

What this paper does is give me pause in accepting the statements that claim AI is magically improving developer productivity.

sketch of GenAI promises vs reality
sketch of GenAI promises vs reality

What would I like to see? I would like papers to look at additional questions not covered in this study:

  • Was the code change a sensible size? Generally, we don’t want bug fixes to be large changes. When bug fixes are large, it introduces new risks.
  • Did the AI generated code fit the idioms used in the project?
  • Was the AI generated code readable?
  • Did the AI avoid adding any new dependencies?

It is early days in our use of GenAI and software development. The models themselves will likely improve (example: context windows), and tool vendors will find ways to improve their tools so the LLM has more information about the codebase. I also know of some developers who are using a BDD style of development coupled with code generation, which might well pay off.

I hope we all learn to be more skeptical of promises about productivity increases without first examining the methodology underneath. (I, for one, have written off the DORA Reports as unreliable.)

Mark Levison

Mark Levison

Mark Levison has been helping Scrum teams and organizations with Agile, Scrum and Kanban style approaches since 2001. From certified scrum master training to custom Agile courses, he has helped well over 8,000 individuals, earning him respect and top rated reviews as one of the pioneers within the industry, as well as a raft of certifications from the ScrumAlliance. Mark has been a speaker at various Agile Conferences for more than 20 years, and is a published Scrum author with eBooks as well as articles on InfoQ.com, ScrumAlliance.org and AgileAlliance.org.

Get Certified

Explore what Scrum is and how to make it work for you in our Scrum Certification training. Hands-on learning will guide you to improve teamwork, deliver quick feedback, and achieve better products and results.

About this course

Focuses on the role of the team and the ScrumMaster. Get the skills and practical experience necessary to improve teamwork, take the exam, and advance your career with a certification that is in high demand today. Often the best fit for anyone new to Scrum.

Learning and Benefits

Relatable Scenarios

Learn on-the-job applications of key Scrum concepts, skills, principles, along with practical solutions that you can apply the next day for difficult, real-life situations.

Respected Certification

Everything you need to earn your Scrum Alliance® ScrumMaster certification, including exam fee and membership, and so much more.

Practical Exercises

With focus on the challenges that real teams face, and tools to dig deeper. You don’t need more boring Scrum theory. You need something you can sink your teeth into to see immediate results.

Jargon-Free Learning

This workshop is not just for software development or people with a computer science degree. We’ve helped many non-software teams with Scrum.

Career Advancement

Use Scrum knowledge to standout at work, get paid more, and impress your customer, all without burning out.

Ongoing Support

Our active Scrum community forum is a safe place to ask questions. Long after you earn the Certified Scrum Master certification, you will have access to the forum, course materials, and additional valuable resources.