The Fork That Actually Stuck
When HashiCorp announced the Business Source License in August 2023, the infrastructure-as-code community didn’t just grumble into Slack channels. They forked. And unlike most open-source drama that fades after six months, this one got serious fast. The Linux Foundation picked up the banner, and OpenTofu became the canonical open-source Terraform continuation. I watched the GitHub star count go from thousands to over 23,000 by January 2026. That’s not niche territory anymore. That’s signal.
But here’s what really matters: OpenTofu isn’t just coasting on licensing outrage. Version 1.9, released in late 2025, shipped with native end-to-end state encryption. This is the kind of feature that matters when you’re actually running production infrastructure. Terraform’s open-source tier doesn’t have it. That’s not a minor distinction. That’s a concrete, ship-able reason to migrate.
The download numbers tell the story too. OpenTofu crossed 4 million weekly downloads by early 2026, up from roughly 1.5 million at launch. That’s not everyone jumping ship, but it’s not a rounding error either. And the trajectory suggests we’re still in the acceleration phase.
What Changed at HashiCorp, and Why It Matters
IBM closed its acquisition of HashiCorp in April 2024 for $6.4 billion. That’s the kind of number that changes incentive structures. When your parent company is a $150+ billion enterprise IT powerhouse, the gravitational force pulls away from scrappy open-source iteration and toward products that enterprises will license.
Terraform 1.10 and subsequent releases have been, to be honest, incremental. Better logging here, some edge-case fixes there. Nothing wrong with that, exactly. Steady maintenance is a valid strategy. But when your competitor is shipping encryption you need, and they’re doing it because they’re still beholden to the open-source community rather than to quarterly revenue targets, the dynamic shifts.
I’m not suggesting IBM is deliberately throttling Terraform development. That’s probably not it. But the incentive alignment is undeniably different now. IBM wants customers on Terraform Cloud, Terraform Enterprise, and their own suite of cloud services. A thriving open-source base that doesn’t need to pay for premium features is fine, as long as it doesn’t become the primary development target. And that’s exactly what’s happening.
OpenTofu Gets Institutional Legitimacy
The CNCF Technical Oversight Committee accepted OpenTofu as a sandbox project in early 2025. I know that sounds like bureaucratic theater, but it’s not. That acceptance is the same pathway Kubernetes and Prometheus used to go from community projects to trusted infrastructure primitives. It means audits, governance transparency, and the kind of institutional blessing that gets enterprises past their legal review boards.
Check the Linux Foundation OpenTofu project page if you want the governance details. The structure is tight. They’ve got TSC representation, clear contribution guidelines, and the kind of transparency that makes security teams comfortable. That matters more than it sounds when you’re making five-year infrastructure bets.
This isn’t credibility theater. It’s the institutional foundation that lets mid-market companies and large enterprises justify picking OpenTofu to their CISOs and architecture review boards. The license is open. The governance is transparent. The leadership is accountable. Those boxes get checked now.
The Migration Wave Is Real
A Pulumi survey of 1,200 platform engineers in January 2026 found that 31% had already migrated at least one environment to OpenTofu. That’s up from 11% the year before. The migration isn’t universal yet, but it’s a third of respondents. For a fork that was literally fighting for its existence two years ago, that’s remarkable.
What’s interesting is the pattern. It’s not “everything or nothing.” It’s selective migration. One environment to test. A staging tier to kick the tires. Then, for teams that have the political capital and technical bandwidth, a gradual march toward full parity. That’s exactly how these transitions actually happen in the real world, not how they happen in the blog posts about them.
The real question is whether the remaining 69% are genuinely content, or whether they’re just on the slower part of the adoption curve. My read: some are happy where they are. But a meaningful chunk are probably waiting for more certainty on the OpenTofu side, or they’re stuck in organizations where tooling changes move glacially. Neither of those things stays static forever.
What This Means for Your Infrastructure Stack
If you’re managing infrastructure at any reasonable scale, you’re probably asking yourself whether to care. Here’s my take: you should care, but not panic. Neither tool is going anywhere tomorrow. Terraform still works. HashiCorp isn’t deleting code. But if you’re starting a new project or making migration decisions over the next 12 months, the calculus is different than it was two years ago.
For teams that care about encryption and prefer open-source governance structures, OpenTofu is now a legitimate first choice. Not just a protest vote. Not just a fallback. A genuine alternative that ships features competitors don’t have. That’s the inflection point we’ve reached.
If you want the deep technical details on what OpenTofu 1.9 actually enables, the OpenTofu official documentation and changelog is comprehensive. It’s the kind of resource that makes it easy to evaluate on technical merit rather than on licensing philosophy.
The fork is no longer hypothetical. It’s a mature, feature-competitive project with institutional backing, governance legitimacy, and real adoption. The question isn’t whether OpenTofu will survive anymore. It’s whether infrastructure teams can justify not at least seriously evaluating it. Based on this year’s migration numbers, plenty already have. What’s your timeline looking like?
