What dFusion AI Labs does

dFusion AI Labs Inc. operates a decentralized AI data and knowledge validation protocol positioned as infrastructure for AI training pipelines. The company's core product, the dFusion Protocol, combines a blockchain-based consensus layer with AI-driven filtering and validation to produce a verified knowledge graph sourced from contributors rather than web scraping. The architecture rests on the Open Fusion Kernel SDK (for node operators to run LLM-based filtering, validation, and voting), AI Data Oracle Filters secured by EigenLayer restaking for on-chain validation and RAG, and Vericore—an AI fact-checking subnet on Bittensor (Subnet 70). Crowdsourced knowledge is ingested through the Genesis Knowledge Program, which has attracted 650K+ contributors producing 80M+ insights with an 18% acceptance rate, yielding a corpus the company describes as 7x larger than Wikipedia. A complementary data layer, the Social Truth DLP (integrated with Vana), ingests 3M+ to 159M+ private Telegram chats via Trusted Execution Environments and rewards contributors in $VFSN tokens.

The commercial surface is anchored by Subnet Slots—NFT-based ownership rights over domain-specific AI data pipelines (12-month lockup, priced in USDC/USDT/BTC/ETH with tiered community discounts up to 25%)—and the Knowledge Pod self-service subnet deployment platform. An Enterprise Master Subscription Agreement (last revised July 15, 2025) provides annual commercial access to data ingestion, subnets, nodes, API access, and model training modules. Decentralized compute is sourced from partners including Aethir, IO.NET, 0G AI, Nesa, Dria, Humans.ai, and others, while validation and RAG infrastructure rely on EigenLayer and Bittensor. The company is incorporated in Seychelles, operates through a Singapore subsidiary (dFusion Technologies Pte. Ltd.) with a Dubai DIFC service address, and is backed by NEAR Protocol, Yuma, Stanford (SBA), 1982.vc, Filter Fund, Unpopular Ventures, and BBF. Headcount is reported at 1-10, led by co-founders Roger Ying (CEO, Stanford MS&E) and Patrick De La Garza (CTO, Columbia CS/Maths).

Name

dFusion AI Labs

Legal name

dFusion AI Labs Inc.

Website

https://dfusion.ai

Company type

Private

Founded year

2023

Operating status

Operating

Headcount range

1–10 employees

Short description

dFusion AI Labs operates a decentralized AI data protocol that combines crowdsourced knowledge (650K+ contributors, 80M+ insights) with blockchain-based AI validation, serving AI developers, node operators, and enterprise customers via Subnet Slot NFTs and an MSA.

Ownership category

akta.pro rank

Product category

Decentralized AI Data Infrastructure

NAICS

Computer Systems Design and Related Services (54151), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182)

SIC

Services-Computer Programming Services (7371), Services-Computer Integrated Systems Design (7373)

akta.pro primary industry

Decentralized GPU/AI Compute Networks (training/inference, GPU renting, model hosting) (FSAPALAC)

akta.pro secondary industries

AI Privacy Engineering & Data Governance (PII, consent, retention) (HDAAAKAB), Model Hosting, Serving & Inference Platforms (HDAAACAB), Responsible AI, Security & Privacy Platforms (Safety, Guardrails, PII) (HDAEANAG)

Keywords

• Decentralized AI protocol
• Knowledge graph validation
• AI data infrastructure
• Blockchain AI data
• Decentralized compute

Headquarters

HQ city

Dubai

HQ country

United Arab Emirates

HQ region

Middle East

Offices2 records

Markets served

GTM type

B2B and B2C

Offering type

Software

Cost components

Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations

Revenue model

1. Protocol Token Economics: Deflationary token model where protocol usage burns tokens. Maximum limited supply with node network rewards and staking rewards distributed to participants. Token staking required to access knowledge graph.
2. Subnet Slot Sales: Sale of Subnet Slots as NFTs representing domain-specific data pipeline ownership. Sold for USDC, USDT, with BTC/ETH acceptance. Subject to 12-month lockup period.
3. Data Access Fees (VFSN Token): Data token ($VFSN) for Social Truth DLP with 28M total supply. 80% of VANA data access fees flow back to $VFSN ecosystem. Token holders need minimum thresholds for data access.
4. Enterprise Subscriptions (MSA): Master Subscription Agreement for commercial access to dFusion Data Protocol including data ingestion tools, subnets, nodes, API access, and model training modules. Annual subscription with 30-day renewal terms.

Pricing tiers

Go-to-market motion2 records

Distribution channels3 records

Marketing channels6 records

Core offering

dFusion AI Labs operates a blockchain-based decentralized AI data protocol that sources and validates high-quality knowledge for AI systems through crowdsourced contributions. Its primary commercial offerings are Subnet Slots — NFTs granting ownership of domain-specific AI intelligence data pipelines — and an Enterprise Master Subscription Agreement for commercial access to its data ingestion tools, subnets, nodes, API access, and model training modules. Supporting products include the Genesis Knowledge ingestion program, the Social Truth Data Liquidity Pool for privacy-preserving Telegram chat contributions, and Vericore, an AI fact-checking service on Bittensor.

Product overview

dFusion AI Labs offers a decentralized AI data protocol platform with multiple integrated modules. The core dFusion Protocol enables blockchain-based knowledge validation and management, augmented by Genesis Knowledge (a crowdsourced knowledge ingestion program), the Social Truth DLP (for Telegram data contribution), and Vericore (AI fact-checking via Subnet 70 on Bittensor). Subnet Slots represent the primary commercial offering, granting NFT-based ownership of domain-specific AI intelligence subnets. The Knowledge Pod module enables private subnet deployment. All components are powered by the Open Fusion Kernel SDK, which node operators use to run AI validation, filtering, and knowledge graph maintenance. The platform's architecture combines decentralized compute infrastructure (via partners like Aethir and io.net) with cryptographic incentives for data quality assurance.

Differentiator

Problem solved

Functional benefit

Brands

• Genesis: Genesis Knowledge Ingestion Program - crowdsourced knowledge contribution platform with over 650K contributors and 80M+ insights.
• Social Truth DLP
• Vericore

Products and services

• dFusion Protocol The core blockchain-based AI data protocol that enables decentralized knowledge validation, subgraph networks for domain-specific AI intelligence, and incentive-aligned data contribution through cryptographic tokens and staking mechanisms. It is the foundational software platform for AI developers, node operators, and enterprise data consumers.
• Subnet Slots NFT-based ownership units granting operators the right to run domain-specific AI intelligence subnets on the dFusion protocol. Operators define data relevance rules, validation standards, and access permissions for specialized knowledge domains and earn rewards for uptime and usage. Sold for USDC, USDT, BTC, or ETH with a 12-month lockup.
• Genesis Knowledge A crowdsourced knowledge ingestion program with 650,000+ contributors and 80 million+ insights, producing a knowledge base 7x larger than Wikipedia with an 18% submission acceptance rate. Contributors earn points redeemable for Subnet Slot discounts and rewards.
• Social Truth DLP A Data Liquidity Pool enabling users to contribute private Telegram chat data securely. Data is encrypted in-browser, pinned to IPFS, and scored in a Trusted Execution Environment. Contributors earn $VFSN tokens while providing real-world behavioral signal data for AI model training.
• Vericore AI fact-checking service implemented as Subnet 70 on Bittensor. Vericore verifies truth through semantic intelligence and decentralized consensus, providing AI-powered fact verification and content validation.
• Knowledge Pod A self-service platform for deploying private subnets. Users set custom incentives and data rubrics, stake rewards to attract contributors, collect continuously flowing domain-specific data, and sell gated access to their subnet or build AI models over 10x more accurate than baseline.
• Open Fusion Kernel SDK An SDK and software stack for node operators to participate in the dFusion network. The kernel handles knowledge contribution ingestion, AI filtering with GPU processing, LLM pre-vetting, knowledge voting and validation, token staking, and knowledge graph maintenance. Community projects can integrate applications with the Open Fusion Kernel.

Quantifiable outcome

• 18% acceptance rate ensures high-quality knowledge contributions
• +1 more outcomes

Segments4 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration6 records

AI capability9 records

Feature7 records

Partnerships

15 partnerships are on record, tiered core, major and minor.

• KaikocoreStrategic or Co-development Partner · 2 July 2025Partnership to develop emotionally intelligent AI agents by integrating Kaiko's EQ+ platform with dFusion's privacy-conscious data layer. Aims to enhance AI's emotional awareness and response accuracy across various applications.
• VanacoreTechnology or IntegrationIntegration via Social Truth DLP enabling user-owned private chat data contribution. Vana's Data Liquidity Pool (DLP) infrastructure with Trusted Execution Environments powers privacy-preserving data contribution. Over 3M chats contributed through this integration.
• AethircoreTechnology or IntegrationDecentralized GPU infrastructure partnership providing high-performance compute backbone for inference, filtering, and validation across all subnets and tasks.
• EigenLayermajorTechnology or IntegrationAI Data Oracle Filters & RAG Optimization secured by EigenLayer's restaking infrastructure, enabling on-chain validation and retrieval-augmented generation.
• VericoremajorTechnology or IntegrationAI fact-checking subnet (Bittensor Subnet 70) that verifies truth through semantic intelligence, building an internet based on meaning rather than clicks.
• TenzrominorOthersDePIN partner in dFusion ecosystem for decentralized infrastructure.
• Koii NetworkminorOthersDePIN partner in dFusion ecosystem.
• NesaminorOthersDePIN/AI infrastructure partner.
• MizzleminorOthersDePIN partner in ecosystem.
• DriaminorOthersAI/compute infrastructure partner.
• IO.NETminorOthersDecentralized compute partner.
• 0G AIminorOthersAI infrastructure partner (QG).
• InferumminorOthersAI/compute infrastructure partner.
• PublicAIminorOthersData/AI partner in ecosystem.
• Humans.aiminorOthersAI infrastructure partner.

Scale indicators3 records

Recent moves6 records

Expansion highlights6 records

Direct peers

• Bittensor: Decentralized AI network of incentivized subnets. dFusion's Vericore product runs natively on Bittensor as Subnet 70, making Bittensor the most directly comparable protocol in terms of architecture, token-incentivized AI subnet design, and target ecosystem.
• Vana: Data Liquidity Pool (DLP) protocol enabling users to contribute private data in exchange for token rewards. dFusion's Social Truth DLP is built directly on Vana's DLP infrastructure and TEE pipeline, sharing an identical privacy-preserving data contribution model.
• PublicAI: Crowdsourced AI training data network paying contributors for labeled data. Competes with dFusion's Genesis Knowledge Program for the same contributor base building human-validated datasets for AI training.
• Gensyn: Decentralized compute protocol for AI model training. Competes for the same DePIN + AI infrastructure narrative and overlaps with dFusion's Aethir/IO.NET compute strategy targeting AI training workloads.
• io.net: Decentralized GPU compute network for AI/ML workloads. Listed as a compute partner of dFusion and competes head-to-head with Aethir for the same decentralized GPU marketplace that underpins dFusion's inference pipeline.

Emerging players

• Aethir: Decentralized GPU cloud for AI inference and gaming. Core infrastructure partner for dFusion's compute backbone; co-founder Mark Rydon is a dFusion advisor. Overlapping enterprise customer base in AI compute creates both partnership and competitive tension.
• Render Network: Decentralized GPU rendering and compute network. Operates in the same DePIN compute category as Aethir and io.net that supply dFusion's inference layer, sharing similar token-incentivized GPU marketplace economics.
• 0G Labs: Decentralized AI infrastructure protocol offering on-chain data availability and compute for AI. Listed as a dFusion partner and competes for the same modular, AI-first blockchain infrastructure narrative.

Broad incumbents

• NEAR Protocol: Layer-1 blockchain with explicit focus on AI and decentralized applications. Backed dFusion financially and ecosystem-wise; competes for the same AI-on-blockchain narrative and provides an alternative chain for AI app deployment.

Others

• EigenLayer: Restaking protocol securing dFusion's AI Data Oracle Filters and RAG optimization (AVS) via EigenLayer's Actively Validated Services. Comparable as the foundational security layer for a new generation of modular AVS, of which dFusion is an early adopter.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks5 records

Key highlights6 records

Customer concentration

• Xx.com/dFusionAI
• GitHubgithub.com/dfusion-ai
• LinkedInlinkedin.com/company/dfusion-ai-labs

Revenue estimate

Valuation estimate

Number of profiles

Profiles3 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about dFusion AI Labs