Peter Steinberger built OpenClaw into one of the most widely adopted open source AI agent projects in the world. Nearly 200,000 GitHub stars. Over 34,000 forks. Developers across Silicon Valley, Beijing, and everywhere in between running autonomous agents on their own machines, with their own models, under their own control.
Yesterday, Sam Altman announced that Steinberger is joining OpenAI. And in the same breath: OpenClaw will live in a foundation as an open source project that OpenAI will continue to support.
This is the right decision. It deserves recognition.
The easy path was not this one#
When a large company acquires a popular open source project, the pattern is predictable. The project goes “source available.” The license changes to something that looks open but is not. Community contributions slow down. Forks appear. Trust erodes. The project that grew because it was open dies because it stopped being open.
We have watched this happen repeatedly. Redis. Elasticsearch. Terraform. Each had valid business reasons for changing their license. Each lost something in the process: the unconditional trust of the developer community that built adoption in the first place.
OpenAI chose differently. They are putting OpenClaw into a foundation. Not “open source with an asterisk.” Not “source available under a non-compete license.” A foundation. The kind of structure that gives the community actual governance, actual continuity, and actual assurance that the project cannot be pulled out from under them.
Why this matters for the ecosystem#
AI agents are becoming infrastructure. They execute real operations: reading files, running commands, making API calls, managing credentials, modifying production systems. The tools that govern how agents interact with the world need to be inspectable, auditable, and trustworthy. Open source is not a nice-to-have for agent infrastructure; it is a prerequisite.
When an agent has access to your AWS keys, your Kubernetes clusters, and your CI/CD pipelines, you need to understand what that agent does. Not what the vendor says it does. What it actually does. You need to read the code. You need to verify the behavior. You need to know that the tool you are trusting with production access is not doing something unexpected behind the scenes.
Closed source agent frameworks ask you to trust the vendor. Open source agent frameworks let you verify. The difference is between behavioral enforcement (the vendor promises to behave) and architectural transparency (you can see for yourself). OpenClaw staying open source means nearly 200,000 developers can continue to verify.
The signal it sends#
OpenAI has been on a complicated journey with open source. The name says “open.” The history has been mixed. Keeping OpenClaw in a foundation is a concrete action, not a press release. It tells the developer community that OpenAI understands something important: the value of an open source project is inseparable from the openness itself. Absorb the project, close the source, and you have acquired a codebase. Keep it open, fund a foundation, and you have invested in an ecosystem.
The distinction matters because ecosystems compound. A foundation-backed open source project attracts contributors who build integrations that attract users who file issues that improve the software that attracts more contributors. This loop works because every participant knows their contributions will remain accessible. The moment that guarantee disappears, the loop breaks.
OpenAI is betting that the ecosystem is more valuable than the control. That bet is correct.
This aligns with how we think about software#
At CPLabs, we believe infrastructure tooling should be transparent. Engineers who trust a platform with their DNS, email authentication, certificates, and compliance posture deserve to understand how that platform works. Open source is not a marketing strategy; it is the foundation of trust between a tool and the people who depend on it.
We are a CNCF Silver Member because we believe in that community and its values. When companies invest in open source genuinely, the entire ecosystem benefits. When they close things off, everyone loses.
Seeing OpenAI make this bet with OpenClaw is encouraging. The more companies that demonstrate commitment to open source, the stronger the case becomes for every company making the same choice. Every time a major player keeps a project open instead of closing it, the ecosystem wins. Every time a project goes into a foundation instead of a proprietary repository, the standard for what “open” means gets reinforced.
Credit where it is due#
It is easy to be cynical about large companies and open source. The pattern of acquire-and-close has been repeated enough times that skepticism is the default response. That is exactly why the exceptions deserve acknowledgment.
OpenAI kept OpenClaw open source. They put it in a foundation. They committed to supporting it. This is the right thing to do, and doing the right thing when the easy thing is different deserves credit.
The open source community is stronger when companies invest in it genuinely, not performatively. OpenAI invested genuinely. We applaud that.
The AI agent ecosystem is young. The norms are still forming. Decisions made now about openness, governance, and community trust will shape the infrastructure that every engineer depends on for the next decade. OpenAI just set a good precedent. The rest of the industry should follow it.
References: