AI-assisted development: Supercharging the open source way

Note: This post is the first in a series that will take readers inside our AI projects, share how we’re equipping our engineers for the AI era, and dig into the legal and ethical frameworks we’re grappling with at Red Hat. Our intention is to not only highlight our practices, but also start discussions and spur ideas on the future of software development.

The world of technology is in the midst of a seismic shift. Artificial intelligence, particularly generative AI (gen AI), is rapidly evolving from a futuristic concept into a tangible tool with the potential to redefine how we build software. For those of us who have built our careers on open source principles of software development, this moment is pivotal. And a bit unnerving. At Red Hat, we've always believed that the most powerful innovations are born from collaboration, transparency, and a shared commitment to solving complex problems. As we head into the AI era, we believe these same principles are not just relevant, but essential.

Using AI in open source isn't about replacing the developer; it's about empowering them. It’s about enhancing the creativity and ingenuity that has always been at the heart of open source. We see AI as a powerful new collaborator in the open source community—a tool that can help us scale software development, tackle more ambitious projects, and accelerate the pace of innovation for everyone. Red Hat is all-in on unlocking the potential of AI, and we intend to bring this to the communities we’re part of.

AI as a collaborator in open source development

For decades, developers have relied on tools to make their work more efficient, from compilers and debuggers to sophisticated IDEs. We see gen AI and AI-powered coding assistants as the next evolution in that toolkit. These are not just novelties. They are practical instruments that can handle tedious, time-consuming tasks, freeing up developers to focus on the complex, creative problem-solving where they add the most value—and, let’s be honest, where many get the most satisfaction from their jobs.

The potential of AI-assisted development allows developers to bring their solutions to life rapidly, reducing boilerplate coding, increasing test coverage, and contributing better documentation. It’s about broadening the community and, ultimately, building better software, faster. While there is a healthy amount of pragmatic skepticism around AI today, we believe the application of this technology is inevitable. Which is why we're focused on ensuring its use has appropriate safeguards and software security mechanisms in place.

Upstream, with AI in the loop

Our journey into AI-assisted software development is guided by the same "upstream first" philosophy that has defined Red Hat for decades. We’re not just adopting these tools internally; we are encouraging and enabling our engineers to use them as they participate in the upstream communities that are the lifeblood of open source.

We expect Red Hatters to be active community members, and using the best available technology is part of that. Using AI coding tools to contribute to upstream projects is a natural extension of how we've always worked. We contribute our code, our expertise, and our passion to the projects we believe in. Now, we will do so with the added velocity that AI provides, while upholding the standards and practices of each unique community.

Navigating the new frontier with openness and trust

Embracing this new technology also means confronting new challenges. How can we guide the provenance and identification of AI-generated code? How do we confirm that it meets the security and quality standards of our communities? What’s also critical: Finding ways to help the maintainers handle the increased load of contributions and manage the AI slop, through the use of AI or other means. These are not easy questions, and we know we don’t have all the answers right now.

What we do have, however, is an unwavering belief that these challenges must be met in the open. The solutions must be developed collaboratively, with the same transparency and peer review that are the hallmarks of open source. Human oversight remains critical. Every line of code, whether written by a human or with the assistance of an AI, must be subject to rigorous review, testing, and validation.

Our commitment is to work with our communities to build the frameworks, establish the best practices, and define the standards for responsibly integrating AI into the open source development lifecycle. As we work with our existing communities and evaluate future projects to participate in, we will aim to provide each with the understanding and tools needed to scale by way of AI.  

The road ahead is long, and there will undoubtedly be challenges. But we are optimistic and committed. We believe that by working together, we can harness the power of AI to amplify the open source way. This post outlines our principles. In the coming weeks and months, we will follow this with a series of posts that detail our practices. We’ll take you inside our projects, share the guidelines we’re providing our own engineers, and discuss the legal and ethical frameworks that underpin our approach. This is a conversation, and we intend to lead it with transparency and urgency. We invite you to join us on this journey as we build the tools, define the standards, and shape the future of software development together.