A Quick Look at Swarm Network, a Decentralized Truth Verification Protocol with $10 Million in Funding

Author: Zen, PANews

Over the past decade, artificial intelligence has moved from academic laboratories into everyday life. The explosive growth of computing power and data has enabled single models to undertake increasingly complex tasks, but has also spawned new challenges—including bottlenecks in large-scale collaboration, credibility issues caused by model "hallucination," and how to ensure system robustness and auditability when the number of intelligent agents increases significantly.

At the same time, blockchain and decentralized technologies, with their immutable and traceable properties, offer a novel approach to resolving trust issues: if information and verification processes can be securely recorded and proven, disputes arising from information asymmetry or unclear data sources can be effectively mitigated. This is where the Swarm Network comes in. Its goal is to organize a large number of automated agents and decentralized human reviewers into a scalable collaborative network. This network uses a hierarchical clustering mechanism to collect and screen information, and then uses cryptographic proof technology to securely store verified claims on-chain.

In short, Swarm not only hopes to be a fact-checking tool, but also attempts to make "trustworthiness" a callable infrastructure, making truth, evidence and credibility assets that can be measured and exchanged in the digital world.

Core Mechanism: Three-layer collaboration of AI agent, human verification, and truth protocol

The core of the Swarm Network lies in the multi-layered collaboration of AI agents and human verification, as well as the "Truth Protocol" based on zero-knowledge proof. Specifically:

• Truth Protocol: Swarm records verified claims on the blockchain, using zero-knowledge proof technology to ensure privacy and security. This protocol allows any verified information to be uploaded to the blockchain, providing auditable verification evidence.

• AI Agent Network: Swarm introduces a large number of automated AI agents that continuously scan and analyze massive data streams—including social media posts, news reports, and market data. These intelligent agents provide a front-end, filtering out information noise and providing preliminary factual assessments, significantly increasing the scale and speed of the verification process.

• Human Verifiers: Complementing the AI agents is a decentralized network of human reviewers. Human verifiers focus on detail and context, assessing the ethical and cultural implications of complex issues and adding depth to AI screening results. Ultimately, this collaborative effort between humans and AI forms a closed loop, ensuring efficient information screening while remaining consistent with human cognition.

• Incentive Mechanism: Swarm issues "Agent Licenses" as proof of participation. Each license is an NFT, allowing holders to operate AI agents and earn network rewards. This tokenized mechanism not only rewards contributors but also encourages honest verification, fostering a self-correcting consensus ecosystem.

Each agent license grants the holder the right to operate an AI agent, and a daily reward mechanism encourages them to contribute computing power and verify results. The project team states that these NFTs are not only a network governance tool but also a pillar for building a decentralized verification economy. Holders can earn income by helping to maintain network integrity. Furthermore, Swarm's collaborations with multiple projects, including Sui, have rapidly expanded its ecosystem, laying the financial and technical foundation for future development.

Together, these elements form Swarm's verification system: AI agents handle large-scale tasks, humans provide in-depth insights, and truth protocols ensure that the final results are public and immutable. Through this three-layer feedback loop of AI, humans, and the blockchain, Swarm rebuilds trust in an information-fragmented environment.

Financing and Team: The AI "Team" Behind $13 Million in Funding

In August 2025, Swarm announced it had raised $13 million in funding. Of this, $10 million came from the public sale of NFT proxy licenses on the Sui network, while only $3 million came from seed round strategic investors, including the Sui Foundation, Ghaf Capital, Brinc, Y2Z, and Zerostage. Sui, Ghaf, and Brinc secured their investment through Swarm's participation in the SuiHub accelerator program in Dubai. According to Swarm's official blog, its seed round was completed in January of this year, led by ZeroStage and Y2Z Ventures.

It's worth noting that a significant number of Swarm Network's founding core team members hail from the AI agent network project Delysium. Swarm founder and CEO Yannick Myson and co-founder and CFO Queena Tsai were both former co-founders of Delysium, leading marketing and business development, respectively. Furthermore, Swarm co-founder and CTO Morty Tabibi was previously the Chief AI Engineer and Product Director at RCT-AI and Delysium. This talent migration brings Swarm a wealth of experience in mature AI agent technology and community operations.

Application scenarios: from social fact-checking to cross-domain data bridging

Swarm is designed to be an infrastructure for information verification and bridging, with potential applications across Web3 and AI:

Social Media and News Fact-Checking: One of Swarm's applications is "Rollup News," an AI-powered fact-checking tool. Users simply flag suspicious content on the platform (e.g., "X"). Swarm's network of agents and verifiers then intervenes, conducting multiple rounds of verification of the information's source and providing cryptographically verified results within minutes. Over 100,000 users have participated, verifying over 3 million posts using the Swarm protocol. This model provides social networks and media platforms with real-time, trusted fact-checking, effectively combating the spread of misinformation and rumors.

Cross-domain data verification and information bridging: Swarm claims to be able to simultaneously connect the traditional internet, cloud services, and various blockchain networks, enabling hybrid verification of multi-source data. For example, Swarm agents can monitor IoT devices and supply chain sensors, automatically triggering smart contracts to record events on-chain when they detect an event (such as the arrival of goods or temperature anomalies). It can also cross-reference on-chain data such as transaction records and NFT information with Web2 data such as news reports and social media content, providing a cross-environmental verification framework. In this way, Swarm helps break down data barriers between the digital and physical worlds, enabling enterprise applications to obtain reliable information inputs.

Trusted AI Systems and Prediction Markets: With AI rapidly developing, Swarm can also provide trust assurance for generative AI and prediction markets. For example, Swarm can be used to verify the authenticity of outputs from generative models like ChatGPT, providing auditable context for AI decisions. Alternatively, in decentralized prediction markets, Swarm-verified facts can be used as reliable market signals.

While these applications are still in the exploratory phase, Swarm’s design paves the way for the introduction of decentralized trusted verification in a wider range of fields. In general, any scenario requiring cross-chain trusted data or decentralized verification can leverage the Swarm network to build a more secure and transparent information infrastructure.

How Multi-Agent Systems (MAS) Achieve Scalable Collaborative Verification

Multi-agent systems (MAS) are becoming a key area of collaborative intelligence, and Swarm-based group collaboration excels in scalability and robustness. By organizing agents into a hierarchical structure of clusters and swarms, the Swarm Network addresses the performance bottlenecks of traditional MAS when hundreds or even thousands of agents collaborate, achieving a balance between local efficiency and global collaboration.

The three core characteristics of MAS—autonomy, collaboration, and decentralization—are the foundation of the system's operation. Swarm adds a fourth element: cluster formation. Clusters enable agents to share local environmental awareness and align on goals. This allows for rapid decision-making, dynamic load balancing, and fault tolerance without requiring awareness of the entire group, significantly improving the system's scalability and privacy.

At the implementation level, Swarm consists of agents with independent goals and decision-making logic. Agents can be organized into "teams" on demand to handle specific tasks. This environment can range from the cloud and APIs to blockchains, smart contracts, and IoT devices, enabling agents to interact across Web2 and Web3 to complete on-chain verification of real-world events and contract triggering. For example, in Rollup News, different agents are responsible for traditional website verification, on-chain timestamp comparison, and cross-chain verification, ultimately collaborating to produce a verification conclusion that can be uploaded to the chain.

Coordination and settlement are ensured by a settlement layer called Truth Protocol. According to the official website, the protocol records every information exchange through a claims-evidence-reputation mechanism, ensuring that all interactions are auditable and tamper-proof. It also achieves scalability through a parallel verification architecture. Task allocation combines auctions, real-time capacity matching, and reputation-based load balancing. Agents bid for tasks based on their historical performance, and smart contracts automatically settle rewards upon task completion and verification, creating a transparent value stream.

Compared to previous centralized or hierarchical control models, Swarm combines decentralized consensus with spontaneously emerging dynamic organizations, enabling the system to automatically reorganize resources and prioritize the most appropriate agents in the face of sudden environmental changes. For developers and operators interested in participating in this ecosystem, holding an Agent License provides early access to privileges such as code-free tools, orchestration rights, and revenue sharing, becoming a key enabler of large-scale collaborative intelligence.