0.mk has zero staff. Can AI run it?

The name was registered in 2009 because it was the shortest URL possible: a zero, a dot, two letters. Seventeen years later the zero means something else. This platform has no staff: no support team, moderators, or on-call engineer. AI handles most day-to-day operations, including the words you are reading right now. A human still owns the service, pays its bills, and can intervene.

Why try this now?

The original 0.mk was a part-time project run by three people. None of them worked on it full time, and it closed in 2014 because the economics did not work. The market was small, while hosting, continued development, spam cleanup, abuse review, and better filtering all cost time and money. This version asks a practical question: can AI absorb enough of that operational work to make a useful free service sustainable where it was not before?

What that actually means

This is not a chatbot bolted onto a normal company. AI performs most of the work that would otherwise require a team:

• It builds the product. New features, bug fixes, deployments to the edge network. The redirect worker, the dashboard, the analytics pipeline: written, tested, and shipped by AI.
• It fights abuse.Every destination URL is scanned against Google's threat intelligence plus our own heuristics. When a phishing campaign targeted the platform, the AI found it in the data, took down the links, blocked the accounts, and wrote the detection rules so it would not happen again.
• It reads your feedback. Reply to any 0.mk email and the message lands in a feedback queue the AI reads, triages, and acts on. Same for the public ideas board, where anyone with an account can propose and upvote features.
• It keeps the receipts.Every decision on the ideas board is recorded publicly with its reasoning. Every moderation action lands in an audit log. Over time this becomes the platform's memory: decisions are not vibes, they are a record the AI learns from.

Where the human fits in

One human owns the domain and infrastructure, pays the bills, reviews the AI's work, and can intervene. A human remains responsible for legal matters, payments, and decisions that cannot be undone. AI handles most routine engineering and operations. That division is the experiment: what does a real service look like when AI is not merely a feature, but its primary operator?

Why trust an AI-run service?

Fair question. Three answers. First, transparency: the ideas board and its decision log are public, so you can see exactly what gets built, what gets declined, and why. Second, preservation: an old backup recovered 657,958 links from the original 2009–2014 service, and those links are useful again. Third, incentives: there is no growth team optimizing for engagement. The operational goal is to keep links fast, safe, and available.

What the platform does today

Short links on the shortest domain, redirecting in under 50 ms from 300+ cities. Smart actions that turn the address bar into a command line: type 0.mk/gpt/your question to open ChatGPT with a prompt, 0.mk/doc/Title for a new Google Doc, 0.mk/wifi/Network/Pass for a WiFi QR code. A prompt gallery where AI prompts get short, shareable URLs. Free developer tools that live at the domain itself. And the boring-but-vital parts: real-time analytics, an API, custom domains, and safety scanning on every link.

Steer it

If you want something built, you do not need to find the owner or send a private pitch. Put it on the board, get votes, and watch the decision get made in public. The zero in 0.mk used to mean short. Now it also means zero staff between an idea and the system evaluating it.