In its earliest days, open source tended to imitate and commoditize high-priced proprietary software (Linux for Unix, JBoss for BEA WebLogic, etc.). Today open source commands the role of innovator, not imitator. From cloud infrastructure and devops automation to machine learning and data engineering, open source projects have become enterprise defaults.
What lies ahead? To see the future of software, it pays to look at what’s happening on GitHub, the world’s largest repository of open source software. We can get a sense of the rising stars of open source—and by extension, industry trends—by measuring GitHub stars, forks, and commit activity. I’ll dig into the data here, but here’s an unsurprising spoiler: AI is the big new thing in open source, just like everywhere else. GitHub is seeing a 98% year-over-year increase in generative AI projects and a 92% rise in Jupyter Notebook usage.
Laying the foundation: Kubernetes and friends
No enterprise can use AI if the rest of its infrastructure isn’t rock solid. For that infrastructure assurance, enterprises for decades have increasingly embraced open source. I mentioned Linux before, but Kubernetes is the past decade’s big winner, with 71% of Fortune 100 companies turning to it as their primary container orchestration tool. That success shows up in 114,000 GitHub stars, more than 40,000 forks, more than 74,000 contributors (from more than 7,800 companies), and more than 314,000 code commits. It’s an amazing example of open source development, and it continues today: There are nearly 2,000 open issues and daily commits in 2025.
Beyond Kubernetes, companies have flocked to open source infrastructure as code (IaC) and container tools to manage cloud deployments. For example, over the past few years HashiCorp Terraform has emerged as a de facto standard for IaC; its popularity shows in its 45,000 stars and 9,800 forks. More recently OpenTofu is giving enterprises another open source option. Even more telling is the rise in IaC usage overall: GitHub’s data shows that declarative configuration languages like HashiCorp’s HCL saw 36% year-over-year growth in 2023. This aligns with a sharp uptick in developers using Terraform and similar tools to standardize cloud deployments. Other cloud-native infrastructure projects, from service meshes (Istio) to monitoring systems (Prometheus), have also grown robust communities.
The use of open source for containerization and pipeline automation exploded in this same timeframe. GitHub’s Octoverse notes that by 2023, 4.3 million repositories on GitHub were using Docker container files, including over one million public repos with Dockerfiles, reflecting the ubiquity of container-based development and deployment. In parallel, infrastructure automation via CI/CD pipelines and “everything as code” grew sharply. Developers are not only containerizing apps but also automating their release processes using tools like GitHub Actions, GitLab CI, and Argo workflows.
One way to interpret this is that open source operational tools are keeping pace with software development. Open source devops projects might not grab headlines like AI does, but they show consistent year-over-year growth in adoption. The rise of HCL and Shell as top languages on GitHub reinforces that ops-focused code is a growing share of open source activity. In practical terms, enterprises are standardizing on these open tools to manage complex cloud environments.
Crucially, these and other cloud infrastructure, containerization, and pipeline automation projects benefit from strong enterprise involvement. Developers from Google, Red Hat, AWS, and VMware contribute code to Kubernetes and related projects, ensuring these tools meet basic enterprise requirements. But it’s more than that. We’re seeing an increasing propensity by competitors to collaborate on common infrastructure platforms, allowing organizations to focus on higher-level innovation.
Machine learning and AI: where the magic happens
Such infrastructure projects make it possible for enterprises to embrace AI. And oh, have they embraced it, thanks to a wealth of open source projects. Open source machine learning libraries and new AI projects experienced unprecedented growth in both usage and community size. Although established frameworks like TensorFlow and PyTorch remained extremely popular, the real story was the advent of generative AI and large language model projects in open source. In 2023 alone, the number of generative AI projects on GitHub jumped 248% year over year, while the count of individual contributors to these projects grew by 148%, according to GitHub data.
Several standout projects exemplify this AI dominance in open source. Hugging Face Transformers, a library unifying state-of-the-art models, soared to over 140,000 stars by 2025, becoming a central tool for natural language processing and model sharing. Newcomer projects saw some of the fastest star-growth ever recorded: LangChain, an AI framework introduced in late 2022 for building applications with language models, rocketed to over 100,000 stars within about a year. Likewise, AutoGPT, an experimental autonomous AI agent, gained more than 174,000 stars, putting it among GitHub’s top projects virtually overnight. Even AI applications like Stable Diffusion (for image generation) and its popular web UI accumulated huge communities. Of course, GitHub stars don’t translate into revenue and are not a hard metric for adoption, but they do indicate just how much AI has captivated the attention of developers.
Indeed, the trend is clear: AI dominates open source growth. By 2023, GitHub noted that some open source generative AI projects were already among the top 10 most popular projects by contributor count. This is a remarkable shift. Historically, the top-contributed projects were often operating systems, databases, or dev tools. Now, AI projects—many started by individual developers or research labs—are rallying huge communities. Again, this AI boom wouldn’t be possible without open source first laying the foundation in other categories. For example, deploying machine learning models at scale requires cloud infrastructure like Kubernetes operators for AI and MLOps pipelines, plus devops automation. Enterprises leveraging AI are simultaneously investing in those open platforms. In essence, AI hasn’t replaced interest in cloud or devops projects, but instead has become their accelerant.
It’s also worth pointing out that although data engineering projects don’t get quite the same attention as generative AI ones, projects like Apache Airflow (increasingly the standard for data pipeline scheduling) or dbt (SQL-centric data transformation) have shown steady growth. These undertakings address complex enterprise needs (e.g., processing millions of records or events), so their adoption tends to be very deliberate. Even so, during the past few years we’ve seen more and more companies opt for open solutions over proprietary ETL/ELT or analytics tools because of benefits such as flexibility and cost.
Time to get on the open source bus
An essential truth has emerged in recent years: Open source is no longer a nice-to-have but a fundamental business imperative. From cloud-native infrastructure and devops tools to AI frameworks and data platforms, enterprises derive significant strategic benefits from active involvement. Particularly with AI going open source so early, open source has become a strategic playbook enterprises need to follow to succeed.