I can hear the objection already: Kai, my stack is PHP-FPM behind nginx, why am I reading about a JavaScript runtime changing hands? Stay with me, because the interesting part of Anthropic buying Bun in December 2025 has very little to do with JavaScript. It has to do with a question every one of us should be able to answer about our own stack: who owns the thing your code runs on, and what do they want from it?

The facts first. Claude Code, Anthropic's coding agent, ships to millions of users as a single standalone binary produced by bun build --compile, no Node.js required on the target machine. That product hit a billion dollars in annual run-rate revenue within six months of its public launch. Bun, meanwhile, had raised 26 million dollars of venture capital and earned nothing at all. Anthropic did the obvious thing and bought its critical dependency outright. Jarred Sumner, Bun's creator, added the detail that makes the whole story land: "The GitHub username with the most merged PRs in Bun's repo is now a Claude Code bot."

Here is my position, stated plainly: this acquisition makes single-vendor runtime ownership the new default for young infrastructure, and it turns neutral governance from a boring procedural detail into a genuine competitive property. The PHP world has been sitting on that property for years without marketing it. We should start.

Let me concede the strongest counter first, because it is strong. On the evidence so far, the deal has been good for Bun users. The team hired engineers a pre-revenue startup never could have paid. Bun 1.3 arrived with a frontend dev server, hot reloading, and built-in PostgreSQL and SQLite clients. Monthly downloads passed 7.2 million, with 25 percent growth in October 2025 alone. Nine months in, the MIT license and the open source commitment have held. Compare that to how foundation-governed projects live: the PHP Foundation funds a handful of core developers and still has to rattle the donation tin every year. If you tell me a deep-pocketed owner with a direct incentive beats chronic underfunding, I get it. When Bun breaks, Claude Code breaks, and a billion-dollar product is a very effective maintenance contract.

But an incentive is not a promise, it is a snapshot. Today Anthropic needs Bun to be an excellent general-purpose runtime, because compatibility with the wider ecosystem is what makes the binary work everywhere. The roadmap already leans toward capabilities for autonomous agent execution, and that is fine while both goals point the same way. The day they diverge, the tiebreaker is not a mailing list vote, it is a product review inside one company. php-src has no equivalent single owner. The RFC process is slow, occasionally exhausting, and produces arguments that run for months over a single function signature. It is also structurally incapable of being steered by anyone's quarterly numbers. I used to file that under inefficiency. I now file it under insurance.

The bot statistic deserves its own paragraph, because it is the part that will reach us next. An AI agent is currently the most prolific merged contributor to the runtime it executes on, with humans reviewing and merging its pull requests. That loop is new, and it will not stay confined to one repository. Picture the moment agent-authored patches start arriving at php-src in volume. Our bottleneck has never been patch supply, it has been review capacity and the trust that reviewers place in a contributor's track record. An ecosystem with a corporate owner can simply staff that review pipeline. Ours has to fund it, and that is a conversation the PHP Foundation should be having before the pull requests show up, not after.

Notice also what the acquisition did not do. Node.js keeps its enormous production footprint, its OpenJS Foundation governance, and two decades of ecosystem inertia, and companies like Netflix, Spotify, and Salesforce adopting Claude Code did not stop deploying on it. Ownership shifts the trajectory of investment, not the installed base. That is exactly why the governance question matters more than the benchmark question: benchmarks change every release, ownership changes maybe once a decade, and you live with the consequences far longer.

So here is what I genuinely want to know from you. If a well-capitalized company offered the PHP project the Bun deal tomorrow, ten funded full-time core developers, real infrastructure money, a public promise to keep everything open, in exchange for one company holding the steering wheel, would you take it? I keep flip-flopping on my own answer, and that tells me the trade is closer than my thesis wants it to be. Tell me where you land.