Perceptron raises $6.5M for decentralized AI data network
Perceptron raised a $6.5 million strategic round to launch a data-questing platform for AI companies seeking community-sourced datasets.
By Dominic Okoye · Staff Writer
· 3 min read
Perceptron raises 6.5M in a strategic funding round to expand its decentralized AI data network and launch a platform where AI companies can request specific datasets from its contributor base. The company did not disclose its valuation, revenue, or the structure of the financing.
The round brought in a long list of Web3 and infrastructure investors, including Sigma Capital, Selini Capital, QCP Capital, P2 Ventures, CoinDCX Ventures, Momentum6, DeFi Capital, Walrus Foundation, Aethir, Colosseum, GuruDev Capital, Tempo Finance, NewTribe Capital, Digital Consensus Fund and CodeCraft Capital.
Perceptron said the new capital will fund the launch of its data-questing platform, improve tools and rewards systems for contributors, and support its plan to grow toward a 5 million-node network. That target remains a company projection, and Perceptron did not give a timeline for reaching it.
What is Perceptron building?
Perceptron is building a decentralized network for collecting, verifying and monetizing data used by AI companies. Its pitch is that AI teams can source real-world datasets through a distributed community instead of relying on centralized scraping systems or one-off data vendors.
The company describes the network as a contributor mesh made up of idle bandwidth, user-supplied datasets and domain expertise. In practical terms, Perceptron wants AI companies to post data requests and receive verified datasets in days, while contributors earn from the data or expertise they provide.
Perceptron said the network has onboarded more than 700,000 nodes. In a separate traction update, the company said it now has more than 300,000 daily active users across more than 807,000 nodes, up from an earlier phase in which live agents reached more than 200,000 users in communities including Telegram and Discord.
Those figures are central to the company’s fundraising case, though Perceptron did not define in detail what counts as a node or how daily active users are measured across those communities. For buyers of AI training and evaluation data, the more relevant test will be whether the network can produce datasets that are accurate, compliant and specific enough to replace existing procurement channels.
How will the new platform work?
The data-questing platform is meant to shift Perceptron from collecting only organically contributed data to taking direct commissions from AI companies. A company seeking a particular type of data would be able to request it from Perceptron’s community, which the startup says can include people with niche knowledge such as doctors, lawyers and native speakers.
Peter Anthony, Perceptron’s UK co-founder and CEO, said the company has already shown it can grow a large node network without paid distribution and that the funding will support the move into commissioned datasets. Nathan Gurr, an investment analyst at P2 Ventures, said Perceptron’s model gives it access to a distributed workforce with specialized expertise and could help it become part of the decentralized AI stack.
The company’s near-term focus is the data-questing launch, with more announcements expected next quarter. Longer term, Perceptron is trying to become a single supply layer for AI data requests, contributor rewards and dataset delivery, a broad ambition in a market where data quality and provenance are becoming more important as model builders look beyond generic web-scale training data.
This story draws on original reporting from Tech.eu.
