Perceptron just closed a $6.5 million strategic funding round, tapping into some of the top Web3 investors to propel its vision of decentralized AI data sharing. The startup is aiming to transform how AI companies source training data by building an open ecosystem where users and AI firms alike can earn rewards for their contributions.
Building a Decentralized AI Data Network
Unlike traditional AI data providers, Perceptron wants to harness the power of a Web3 community to generate and share datasets. Their platform incentivizes participants through tokenized rewards that accumulate points for activities, pointing toward a future airdrop. However, the team cautions users about counterfeit NFT snapshot attempts, highlighting the risks in emerging token economies.
Co-founder and CEO Peter Anthony highlighted the company’s rapid growth, noting how their decentralized network organically scaled to hundreds of thousands of nodes. This milestone signals strong community engagement and a promising foundation to launch the new data-questing platform. This tool will enable AI companies to precisely commission high-value data directly from the network’s contributors, creating a more efficient pipeline than traditional methods.
Backing from Industry Leaders and Next Steps
The $6.5 million round includes support from Sigma Capital, Selini Capital, among other Web3-focused investors and trading firms. This marks Perceptron’s largest funding to date, following the fusion of Blockmesh with Perceptron Network in mid-2025 that laid the groundwork for their current infrastructure.
With these funds, Perceptron plans to accelerate product development and expand its tooling to foster a vibrant AI data marketplace. This could reshape the space of AI training, making it more accessible and community-driven. The project’s progress echoes other moves in blockchain and AI convergence, not unlike innovations seen in the crypto API space for AI agents.
This material is for informational purposes only and does not constitute financial advice.
