# dfpn

> dfpn is a decentralized coordination layer for deepfake detection on Solana. Independent operators run their own detection models on their own GPUs; a commit-reveal protocol stops collusion; stake, rewards, and slashing in the DFPN SPL token keep operators honest; and every verdict is anchored on-chain with a full audit trail.

dfpn is operated as an open protocol. Independent node operators compete to produce accurate verdicts on images, video, and audio. Clients submit a media hash and a fee; multiple workers analyze it independently; a reputation-weighted consensus produces the verdict. The code is MIT-licensed; the on-chain programs are Anchor 0.30.1; the worker is Rust; the SDK is TypeScript; detection models run as Python subprocesses or HTTP services on operator hardware.

## Docs

- [Home](https://dfpn.cryptuon.com/): Overview, the problem, the lifecycle of a verdict, and roles
- [About](https://dfpn.cryptuon.com/about/): What dfpn is, what it explicitly is not, the threat model, and project philosophy
- [Tokenomics](https://dfpn.cryptuon.com/tokenomics/): Supply, allocations, emissions, fee splits, stake floors, slashing schedule, scoring weights, governance
- [Field notes](https://dfpn.cryptuon.com/blog/): Writing on detection, decentralization, threat modeling, and provenance

## Compare

- [vs C2PA](https://dfpn.cryptuon.com/compare/c2pa/): Open standard for content provenance; complementary, not competing
- [vs Truepic](https://dfpn.cryptuon.com/compare/truepic/): Commercial provenance camera SDK; complementary, not competing

## Optional

- [RSS](https://dfpn.cryptuon.com/rss.xml): Field notes feed
- [GitHub](https://github.com/cryptuon/dfpn): Source code, threat model, and tokenomics doc