> For the complete documentation index, see [llms.txt](https://docs.bonzai.sh/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.bonzai.sh/ownership-and-network/bonzai-utility.md).

# Why $BONZAI Exists

`$BONZAI` gives private AI work a shared economic standard when it becomes publishable, reusable, or revenue-producing.

A locally downloaded model can be private, but it cannot by itself answer:

* Who supplied the training data?
* Who curated and evaluated it?
* Who trained, served, or maintained the model?
* Who owns the companion or generation?
* Who should earn when a derivative asset becomes useful?

BonzAI combines two things:

1. **Proof of Contribution** records useful work and immutable ownership shares.
2. **Staked `$BONZAI`** records durable commitment and creates additive utility weight.

```
useful contribution + staked $BONZAI
→ trusted publishing, routing, reputation, and additional rewards
```

## Current Desktop utility

Desktop provides:

* staking, lock extension, contribution score, tier, and utility weight;
* contribution packs and provenance-aware training;
* immutable asset revenue routes;
* validated companion, fine-tuned, and abliterated model economies;
* provider, companion, job, skill, and content reward paths;
* Markets for issued companion/model tokens.

## Why staking is not a toll

BonzAI does not require a stake for ordinary local chat or generation. Staking becomes relevant when a participant asks the shared economy to trust, rank, publish, route, or reward an asset.

## Why stake alone is insufficient

Stake does not prove that someone improved a dataset or model. Contribution scores must be evidence-backed and attested by an authorized score oracle. Base contributor royalties remain fixed after the route receives revenue.

## Utility weight

The staking vault calculates:

```
utility weight = stake × lock multiplier × contribution multiplier
```

Utility weight can be used by publishing, curation, provider routing, and the separate staking bonus. It does not rewrite the base contribution split.

## The moat

```
BonzAI+ captures attributed data
→ Desktop curates, trains, validates, and generates
→ models, companions, jobs, and media preserve origins
→ markets and providers create usage
→ revenue returns to recorded contributors
→ staking adds trust and long-term alignment
```

This loop makes `$BONZAI` useful because local AI assets need coordination, not because prompts need another payment token.