Quoting Andreas Kling 引用Andreas Kling

The open-source dream just hit its first real AI-era wall. Ladybird, the ambitious independent browser project, has officially slammed the door on public pull requests. The reason? The flood of low-effort, AI-generated code submissions has poisoned the well of good faith. This isn't just a policy change for one project; it's a canary in the coal mine for the entire open-source ecosystem.

For decades, the social contract was simple: you see a bug, you write a patch, you submit it. The time and sweat poured into that patch was the currency of trust. As project creator Andreas Kling puts it, "A substantial patch used to imply substantial effort, and that effort was a reasonable proxy for good faith." That proxy is now worthless. When an AI can conjure a plausible-looking diff in seconds, the old signal of human diligence has been permanently jammed.

And here’s the brutal, unpopular truth: this isn't primarily about authenticity. Kling is right to say "whether code was typed by hand is beside the point." The core problem is accountability. A patch whipped up by a contributor who barely understands the codebase isn't just potentially buggy; it's an act of profound irresponsibility. When that code breaks a user's bank login or corrupts their data, it’s the project maintainers—Kling and his team—who will be on the hook. They are the ones who must triage the bug reports, perform the forensic debugging, and answer to the community. Allowing anonymous, AI-hurled darts at their codebase is an unsustainable liability.

This move exposes a dirty secret about the "hacker ethos" we’ve long romanticized. We celebrated the lone contributor fixing a bug from a café, but we systemically undervalued the thankless labor of maintenance and review. AI hasn't created this problem; it has weaponized it, making it trivially easy for well-meaning but lazy people (and outright grifters) to become a net drain on a project's health. The volume alone is a denial-of-service attack on maintainer bandwidth.

Is it a tragedy for open-source collaboration? Yes, a genuine one. There’s a unique magic in lowering the barrier to entry, in allowing the kitchen sink programmer to eventually grow into a core maintainer. But that model was built on a foundation of scarce human attention. We are no longer living in that world. The new, harsh reality is that access must be earned. For Ladybird, this likely means a shift to a curated team of known contributors with a proven stake in the project. It’s a return to a more classic, guild-like model—a fortress with a moat, rather than an open bazaar.

Critics will call this elitist, a betrayal of open-source principles. But they’re confusing the principle with the mechanism. The principle is building great, public software together. If the mechanism of open pull requests becomes a vector for sabotage and unsustainable load, then changing the mechanism isn't a betrayal—it's an act of survival. Other major projects will follow. We will see more gated contribution processes, more signed CLAs, and more projects defaulting to "discussions welcome, but send patches only via our trusted core team."

We are witnessing the end of the innocent era of open source. The era where "anyone can contribute" was an unalloyed good. It’s being replaced by a more sober, more corporate-feeling paradigm of verified contribution. It’s less exciting, perhaps less democratic, but it might be the only way to ensure that the critical software we depend on isn’t quietly hollowed out by an avalanche of machine-generated garbage. Ladybird isn't closing itself off; it's growing up. And that maturation process, while painful, is what happens when a hobby project starts to matter for real.

开源梦想首次遭遇AI时代的真正壁垒。雄心勃勃的独立浏览器项目Ladybird已正式关闭公开拉取请求。原因何在？大量低质量、AI生成的代码提交毒化了开源社区的善意之井。这不仅是某个项目的政策变更，更是整个开源生态系统的预警信号。

开源梦想首次遭遇AI时代的真正壁垒。雄心勃勃的独立浏览器项目Ladybird已正式关闭公开拉取请求。原因何在？大量低质量、AI生成的代码提交毒化了开源社区的善意之井。这不仅是某个项目的政策变更，更是整个开源生态系统的预警信号。

数十年来，开源社区的社交契约简单明了：发现漏洞，编写补丁，提交代码。投入补丁的时间与汗水曾是信任的货币。正如项目创始人安德烈亚斯·克林所言："一个实质性的补丁曾意味着实质性的努力，而这种努力曾是善意的合理象征。"如今这种象征已失去价值。当AI能在数秒内生成看似合理的代码差异时，人类勤勉的旧有信号已被永久干扰。

而这里存在一个残酷且不讨喜的事实：问题核心并非代码的真实性。克林说得对——"代码是手打还是机器生成并不重要"。真正的核心在于责任归属。一个由几乎不懂代码库的贡献者草草完成的补丁，不仅可能充满漏洞，更是极度不负责任的行为。当这些代码破坏用户的银行登录系统或损毁其数据时，项目维护者——克林及其团队——将首当其冲。他们需要处理漏洞报告、进行故障排查调试、对社区做出解释。允许匿名AI随意攻击代码库，是一种难以承受的责任风险。

此举揭露了我们长期浪漫化的"黑客精神"中不堪的秘密。我们歌颂独自在咖啡馆修复漏洞的贡献者，却系统性地低估了维护与审查工作的默默付出。AI并未创造这个问题，而是将其武器化，让善意但懒惰的人（乃至纯粹的投机者）极易成为项目健康的净消耗者。仅其提交量就已构成对维护者精力的分布式拒绝服务攻击。

这对开源协作而言是否意味着悲剧？是的，一个真实的悲剧。