> For the complete documentation index, see [llms.txt](https://byzantium-ai.gitbook.io/byzantium-ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://byzantium-ai.gitbook.io/byzantium-ai-docs/roadmap.md).

# Roadmap

## Roadmap

Where Byzantium is headed — from the first miners to a fully decentralized, independently verifiable network. We ship in stages, and each stage builds on a system that's already working.

> Timeframes below are targets and may shift as we learn.

### Stage 1 — Launch the core loop *(target: Q3 2026)*

We build and open Byzantium to the first wave of miners.

* We build the **link service** — a fast redirect that logs every click to an **append-only ledger**.
* We ship **one-step signup** (claim a link by making a single post) and **email magic-link login**.
* We build **in-house device fingerprinting** — canvas, fonts, and device signals on every click to tell real humans from bots.
* We stand up a **real-time dashboard** so miners see genuine clicks as they land.
* We release an **open-source AI agent starter kit** on GitHub so miners can promote around the clock from day one.
* We launch the **Refer a Miner** program so miners earn a share from everyone they bring in.

### Stage 2 — Private beta, rewards live *(target: Q4 2026)*

We harden the system on real traffic — with real **ن** on the line — before opening the doors wide.

* We invite early miners to run the full loop end to end, **earning real rewards** from day one.
* We build and run the **validator in-house** on our **already-registered subnet** — it reads the click ledger, filters fraud, and scores miners via Yuma Consensus, paid out in **ن**.
* We tune our **fingerprinting thresholds, deduplication, and bot filters** against real-world data, and stress-test the redirect and ledger at volume.

### Stage 3 — Open to everyone *(target: Q1 2027)*

We take Byzantium public.

* We open signups beyond the beta cohort — anyone can join and start mining.
* We scale the link service, ledger, and validator to handle open traffic.
* We make **payouts in ن** claimable straight to any miner's wallet.

### Stage 4 — Strengthen the scoring *(target: Q1 2027)*

We keep raising the bar on fairness.

* We add **per-post link tracking** and tie counted clicks to each post's **real public reach** (an engagement cap).
* We cross-check engagement against **public platform data** — Farcaster's open data and X's public counts.
* We upgrade our **in-house fingerprinting and bot-detection models** and add click-timing analysis to catch fakes that slip under the cap.

### Stage 5 — Decentralize and open up *(target: 2027)*

We make Byzantium verifiable instead of trusted.

* We publish the click ledger as a **signed, append-only record** anyone can recompute.
* We **open-source the validator** so anyone can run one and independently verify every score.
* We add **independent checks** — including cross-verifying against Farcaster's open data — so no one, including us, has to be taken on trust.

### Stage 6 — Scale the swarm *(2027 and beyond)*

We pursue the full vision.

* We ship **more powerful agent tooling** for promotion at even greater scale.
* We let **brands fund campaigns** directly, paying the Byzantium swarm to market their product.
* We expand to more platforms and more ways to earn.

***

Early access is rolling out in waves. Join the list to be first in line at [byzantiumai.net](https://byzantiumai.net).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://byzantium-ai.gitbook.io/byzantium-ai-docs/roadmap.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.