React leaves Meta and runs under its own foundation
I have been writing React code for about a decade, and I have
watched the project go through three identity shifts. The first was
the BSD + Patents grant (a license that gave users broad rights but
included a patent grant that terminated if you sued Facebook), which
the community hated and Facebook eventually walked back. The second
was the move to MIT, which bought the project five good years of
trust. The third, which landed in early 2026, is React moving out
from under Meta entirely and into a new independent foundation. I
spent the last few weeks reading the governance docs, talking to a
few people close to the move, and updating my own client projects
to think about the change. Here is what is real, what is marketing,
and what the move means in practice.
The short version is that the foundation move is a real governance
change, not a rebrand. React, React Native, and JSX (a JavaScript
syntax extension that lets you write HTML-like code inside
JavaScript) are now owned by the React Foundation, a new
non-profit (a tax-exempt organization structured to serve a public
purpose rather than to generate profit for owners) with its own
board, its own IP (intellectual property, the legal rights to the
code and trademarks) assignment, and its own technical steering
committee. Meta contributes code, contributes engineers, and
contributes funding, but does not control the project the way it
did when the React team was an internal Meta org. That is the part
that matters. The rest of this article is about what the change
means for the people who actually ship React code, and for the
people who are tired of betting their frontend (the part of a web
app that runs in the user’s browser) on a single vendor’s whim.
What changed in the governance
For the first time in React’s life, the project has a published
governance document that is not “Meta decides.” The technical
steering committee has nine seats. Three are filled by Meta
engineers (down from effectively all of them). Three are filled by
external maintainers elected by the contributor community. Three
are filled by representatives of foundation member companies, who
are voted in by the members. Major decisions, including the React
release cadence, the roadmap, and any breaking changes to the
public API, require a 6-of-9 vote. That is a higher bar than a
single corporate sponsor. It is also a higher bar than most
open-source projects bother to set.
The IP assignment is the part I was most curious about. The
foundation now holds the copyright and trademark for React, React
Native, and JSX. Meta has granted the foundation a perpetual,
irrevocable license to use the React trademark and the React code,
but the foundation is the legal owner. If Meta walked away
tomorrow, the foundation would continue to own and ship React.
That is the part the move is really about. It is not about Meta
giving up influence (Meta is still the largest contributor and
funds a quarter of the foundation’s budget). It is about
codifying a path where the project does not depend on Meta’s
continued engagement.
What did not change
A few things I want to be clear about, because the marketing around
the move overstates what changed.
First, Meta is still the dominant contributor. The React team at
Meta is still a 60-person org. Most of the code in the next major
release is still written by Meta engineers. The foundation is real,
but the contributor base is still concentrated. If you are betting
on a fully community-driven React, you are going to be
disappointed for at least a year or two.
Second, the license is still MIT. The foundation did not relicense
to AGPL, did not add a CLA (Contributor License Agreement, a
contract that says the foundation can use your contributions in any
way it wants) with assignment, did not change the patent posture.
The license is the same license that has been in place since 2017.
For most users, that is the right call. For some enterprise users
who were hoping for a license change, it is a miss.
Third, the release cadence is unchanged. The 18-month major version
cycle is still the plan. The next major release, React 20, is
still on the same roadmap. The foundation inherited the roadmap,
not a new one.
What this means if you ship React
For most people who ship React in production, the change is a
quiet positive. You can now point to a governance document that
says the project will continue even if the founding sponsor
leaves. That is useful in two cases. First, if you are a CTO (chief
technology officer, the executive who owns the technical decisions
at a company) trying to justify a frontend stack to a conservative
CIO (chief information officer, the executive who owns the IT
budget and risk posture), the foundation is a tangible answer to
the “what if Facebook kills React” question. Second, if you are a
startup whose entire product is built on React Native, the
foundation is insurance against the platform risk (the risk that a
platform you depend on changes direction, raises prices, or shuts
down) of depending on a single vendor’s mobile strategy.
For a smaller group of people, the change is a tactical opening.
If you are a contributor, the new governance makes it easier to
get elected to the steering committee. The community seats are
real, and they are filled by a vote of contributors with a track
record of merged PRs (pull requests, proposed code changes). If
you have been meaning to get more involved in React, the
foundation era is the time to do it. If you are a company that
depends on React, the foundation has a member program that gives
you a seat on the member council. The annual dues are not
published, but a few people I talked to said the starting tier is
$50,000 per year. That is real money, and it is the kind of money
that gets you a seat at the table for roadmap decisions.
What I would tell past me
If I could send a message back to the version of me that started
shipping React in 2014, I would say three things.
- The platform risk was always real, and the foundation makes it smaller but does not eliminate it. A foundation can still lose contributors. A foundation can still go stale. The foundation model is better than the single-vendor model, not a guarantee. Plan accordingly. Keep your options open on the framework side (the people who got burned by Angular’s breaking changes in 2016 are still twitching). Keep your options open on the runtime side (the people who got burned by Node’s API churn in 2018 are also still twitching).
- Vendor moves are usually about the next round of growth, not the current one. Meta did not move React to a foundation because of community pressure. Meta moved React to a foundation because Meta’s investment in React is now a fraction of its investment in AI infrastructure (Meta is spending more on Llama training in a quarter than it spent on React in a decade), and the governance overhead of running a community project is not worth it for a strategic non-priority. Read the move as a signal that React is mature and Meta is moving on. That is good news for the project. It is not the news some people wanted to hear.
- The license is the thing that matters, and the license did not change. A foundation can relicense. A foundation can add a CLA. A foundation can change the patent posture. None of those happened, and any of them could still happen. Watch the governance minutes, not the press releases.
- The contributor seats are real, and the window to influence them is open. If you have been waiting to get more involved in React, the foundation era is the time. The first community-seat election is scheduled for late 2026, and the seats will be filled in early 2027. Contributing now is investing in influence later.
Trade-offs
Running a foundation is not free. The foundation has overhead: a
board, a budget, a legal structure, a member program. The
overhead is paid for by member dues and by Meta’s continuing
contribution. In our case, the immediate effect is small. The
foundation has hired a small staff, has published the governance
docs, and is setting up the election process for the community
seats. None of that changes the day-to-day experience of shipping
React code. Your math will be different if you are a contributor
waiting for the community seats to be filled (the first election
is scheduled for late 2026, and the seats will be filled in
early 2027) or a company trying to get a seat on the member
council (dues start at $50,000 per year).
The foundation does not eliminate platform risk. The same risks
exist, just distributed across more shoulders. The biggest
remaining risk is contributor concentration: most of the code is
still written by Meta engineers, and if Meta’s React team shrinks
faster than the foundation can backfill, the release cadence will
slow. That is a real risk, but it is a smaller risk than the
single-vendor version.
The license is unchanged. That is a deliberate choice, and it is
the right choice for most users. If you were hoping for AGPL or
for a CLA with explicit patent grants, you will be waiting.
If you ship React in production, the move is a quiet positive
that does not require any code changes. If you are a contributor,
the move is an opening that is worth taking. If you are a
company that depends on React, the move is a chance to buy a
seat on the council for a known price.
Bottom line
React is now under a real foundation with real governance, not a
vendor-controlled project. The license is the same. The roadmap
is the same. The contributor base is still concentrated at Meta.
The change is real but modest, and for most people the right
response is to note it and move on. For a smaller group of
contributors, companies, and CTOs, it is a tactical opening
worth taking.