Nathen Harvey, DORA Lead at Google Cloud, delves into how generative AI is rapidly reshaping the software development landscape, presenting both exciting opportunities and complex challenges.

Harvey reflects on his own journey, from early web operations and his time at Chef to leading the DORA program after its acquisition by Google. DORA (DevOps Research and Assessment), now in its 10th year, continues to be a definitive source for understanding how technical teams improve performance and value delivery. The program’s annual reports, especially its latest focused on generative AI, offer research-backed insights that are widely used across the industry.

The conversation dives deep into this year’s findings, starting with the rapid and widespread adoption of AI. Nearly 90% of organizations surveyed are prioritizing AI from a top-down perspective, while over 76% of practitioners report already using AI tools daily. This simultaneous push from leadership and uptake by developers indicates that AI isn’t a passing trend—it’s a permanent shift.

But the data reveals nuance. While AI tools are helping improve areas like documentation, productivity and developer satisfaction, there’s also a measurable decrease in software delivery stability and throughput. Harvey suggests that this may be due to larger and more complex change sets generated by AI tools, which clash with workflows optimized for smaller, more iterative updates. In essence, we’re accelerating one part of the pipeline while other parts can’t yet keep up.

Harvey emphasizes the importance of interpreting DORA findings in context—using them as hypotheses to test within each unique organization. He also raises critical questions about workforce development, particularly how AI might reduce opportunities for junior developers to gain real-world experience and grow into senior roles. The future, he notes, will require balance: leveraging AI as an amplifier while still investing in people and process.

The takeaway? AI is transforming DevOps, but not without growing pains. To navigate this shift, organizations need to blend data-driven research with community collaboration, continuous experimentation, and a commitment to developing talent at every level.