dFusion AI

AI performance now depends as much on knowledge quality as model capability.

Yet much of the information that matters most still lives beyond public search, scattered across private chats and specialized data sources that conventional search barely reaches.

dFusion AI was created to unlock that layer ⚡

The protocol is built around an open knowledge graph for the Agentic Web, where contributors submit knowledge, the system filters and validates it, and accepted information becomes part of a shared layer designed for AI systems, enterprise workflows, and AI agents.

That core layer already extends into live products. Vericore runs on Bittensor Subnet 70 as the semantic fact-checking layer, while the Social Truth DLP on Vana turns Telegram data into a community-owned dataset. Aethir supports the GPU compute layer across the subnet stack.

🪂 Testnet, Genesis Points, and $VFSN Rewards

The dFusion testnet is live, and it is the main public route into the ecosystem. Through Fusion Intelligence, users can search across mainnet and subnets, upload content for extraction and analysis, and earn Genesis points through accepted contributions. dFusion has also confirmed that at least 1% of future token supply is reserved for Genesis.

There is a second reward track through the Social Truth DLP.

This route rewards contributors in vFusion ($VFSN) for encrypted Telegram data processed through Vana’s Data Liquidity Pool model. The official page lists 28,000,000 total $VFSN, with 21,000,000 intended for contributor rewards over time.