The Fork That Actually Matters

I need to be direct: we’re watching something genuinely consequential unfold in the infrastructure-as-code space. When HashiCorp pivoted Terraform to the Business Source License in August 2023, most people assumed the Linux Foundation fork would become a well-intentioned but ultimately irrelevant artifact. Those people were wrong. OpenTofu isn’t just surviving—it’s competing in ways that are forcing real technical and business decisions across organizations that have invested heavily in Terraform for years.

The shift became undeniable in late 2025 when OpenTofu released version 1.9 with native end-to-end state encryption built directly into the open-source distribution. This is the kind of feature that matters in the real world. Your infrastructure state file contains secrets, credentials, and sensitive configuration data. Terraform’s open-source tier has always left encryption as an afterthought or a third-party problem. OpenTofu solved it natively. That’s not a marginal improvement. That’s a legitimate security advantage.

Numbers That Reflect a Real Shift

The adoption metrics have stopped being theoretical. The OpenTofu GitHub repository crossed 23,000 stars by early 2026, with over 4 million weekly downloads reported on their public transparency dashboard. That’s a far cry from the roughly 1.5 million weekly downloads at launch. Three times the download volume in under two years for a fork that conventional wisdom said would fade. When you’re seeing that kind of growth trajectory in infrastructure tooling, you’re not looking at momentum. You’re looking at migration.

Perhaps more telling is what happened in the enterprise survey space. Pulumi’s State of Infrastructure as Code survey in January 2026 polled 1,200 platform engineers and found that 31% had already migrated at least one environment from Terraform to OpenTofu. That’s a threefold increase from the 11% in the prior year’s survey. We’re past the early adopter phase. We’re in the “mainstream consideration” phase, which means your organization probably has at least one team thinking about this decision right now, even if you don’t realize it.

Why Terraform’s Momentum Actually Stopped

Here’s the uncomfortable truth for HashiCorp: IBM’s acquisition in April 2024 for $6.4 billion didn’t energize the Terraform open-source community. It scared it. And understandably so. IBM historically optimizes portfolios for enterprise value, not community satisfaction. Terraform 1.10 and its subsequent releases have been measurably incremental. The innovation pace slowed. Features that might have landed in quarterly releases now appear to require boardroom sign-offs.

I’m not saying this to be provocative. I’m saying this because I’ve watched this movie before with other communities. When a major acquisition happens and your open-source project suddenly needs to justify its existence to a vendor’s earnings calls, the open-source side of the house loses prioritization. The community didn’t imagine this. They experienced it. The Terraform core maintainers are as talented as they ever were. The issue isn’t talent. The issue is corporate architecture deciding that OpenTofu’s competitive threat isn’t worth the engineering investment to crush it decisively.

The Legitimacy Inflection Point

What really shifted the conversation was the CNCF’s Technical Oversight Committee accepting OpenTofu as a sandbox project in early 2025. That’s not ceremonial. That’s governance infrastructure. OpenTofu now occupies the same legitimacy pathway that Kubernetes, Prometheus, and etcd used to build enterprise trust. When the Linux Foundation’s governance stamp is on your project, procurement departments stop asking “is this a real project?” and start asking “why aren’t we using this?”

This matters because infrastructure tooling adoption is sticky precisely because it’s hard to switch. Nobody wants to migrate their entire infrastructure state from one tool to another. But the calculations have shifted when the alternative is backed by the Linux Foundation, has CNCF governance oversight, and is demonstrably more secure in areas where your current tooling leaves you exposed. OpenTofu official documentation and changelog shows a project iterating confidently on features that matter to operators, not features that look good in quarterly business reviews.

What This Means for Your Decision

If you’re running Terraform at scale, you’re not making a mistake by staying with Terraform. The tool is mature, well-documented, and integrated into countless workflows. But you should acknowledge what’s happening: you’re now choosing to stay with Terraform despite OpenTofu’s advantages, not because Terraform is obviously superior. That’s a meaningful distinction. Terraform’s position is now defensible rather than inevitable.

For teams building new infrastructure platforms, the calculation is different. You’re not inheriting architectural debt. You get to choose between a tool that’s incrementally maintained within a megacorp’s portfolio and a tool that’s actively developed by a community that has genuine skin in the game. The Linux Foundation OpenTofu project page reflects an organization treating this as a long-term commitment, not a corporate asset to be optimized.

The fork is real competition now. That’s actually good news. Competition drives innovation. If Terraform feels more pressure, HashiCorp will almost certainly accelerate feature delivery on the open-source tier. If OpenTofu keeps capturing adoption, the Linux Foundation will continue investing in governance and stability. Infrastructure teams get better tooling either way. The only thing that changed is that you finally have to think about which one you’re using instead of assuming the choice was already made for you. I got excited about infrastructure tooling for the first time in five years. That should tell you something.