Perceptron Migrates 800,000-Node Intelligence Network to Walrus
Perceptron is integrating Walrus as the storage layer for its enterprise data suite, giving hedge funds, AI labs, and enterprise buyers cryptographic proof of data integrity.
4 min read
Walrus’ verifiability and programmable access lay the foundation for Perceptron's next generation of enterprise data products.
Perceptron is a distributed intelligence infrastructure company that lets anyone contribute to the future of AI. Its network combines autonomous agents, human contributors, and AI-powered data processing to collect, enrich, and distribute high-value datasets for AI developers and enterprises. Perceptron is integrating Walrus as the storage layer for its enterprise data suite, giving every dataset a verifiable identifier and immutable history so customers can independently confirm its integrity.
Key takeaways
-
Perceptron integrates Walrus as the storage layer for its enterprise data suite, bringing data provenance to its network of more than 800,000 contributor nodes.
-
Every dataset stored on Walrus is assigned a verifiable identifier, giving Perceptron's enterprise customers cryptographic proof of integrity.
-
Walrus provides the backbone for Perceptron's upcoming Data Questing Platform and marketplace for permissioned, verifiable real-time data products.
Perceptron, a distributed intelligence infrastructure company that lets anyone contribute to the future of AI, is integrating Walrus as its storage layer for its enterprise data suite. The integration brings verifiable data provenance and scalable storage to a platform that spans more than 800,000 nodes globally and serves clients including Aethir, Kaiko AI, Hub.xyz, and Everlyn AI.
Why Perceptron chose Walrus
To date, Perceptron has relied on centralized storage providers. This surfaced limitations around data provenance, a recurring theme in enterprise data conversations, particularly in sectors with data compliance and sourcing requirements.
Walrus addresses this by giving each dataset a verifiable identifier and immutable history, allowing hedge funds, quant researchers, and enterprise buyers to verify data’s integrity.
Walrus also avoids the pricing dynamics of centralized cloud storage, where data accumulation drives compounding costs and outbound data movement triggers egress fees. Both have created friction for Perceptron as the network has scaled.
Peter Anthony, Co-Founder and CEO of Perceptron, said: "A decentralized acquisition layer needs a decentralized storage layer – anything else is a structural mismatch. Walrus gives us verifiable data provenance, which is exactly what enterprise buyers need when they're making decisions based on our signal."
Underpinning the next phase of enterprise data
Today, Perceptron's contributors share idle bandwidth that the network uses to collect publicly available data. The Data Questing Platform, launching as part of Perceptron's upcoming roadmap, will let contributors take on active tasks: annotating images, labeling datasets, and submitting structured data directly to the network.
This expanded supply of intelligence will meet the growing demand for structured, verifiable, real-time data, particularly amongst enterprise buyers operating under sourcing and compliance scrutiny. Walrus is the storage layer Perceptron will scale into that demand, with permissioned access and verifiable provenance built in.
Kimberly Logan, Head of Product at Walrus Foundation, said: "Data verifiability is becoming one of the newest competitive advantages for businesses serving enterprise customers. Hedge funds, quantitative researchers, and AI companies depend on real-time intelligence and they increasingly need to prove the integrity of that data. The next frontier of AI will be won on data, and as agentic systems demand proof over trust, the verifiability Walrus brings becomes essential.”
Next: A data marketplace
Walrus integration also positions Perceptron to operate a marketplace where every high-value, real-time dataset carries proof of where it came from. As enterprises increasingly rely on proprietary data to train and operate AI systems, verifiable provenance becomes part of the product itself.
Learn more about Perceptron, and explore how Walrus makes data verifiable for AI and onchain finance.
FAQs
What changes for Perceptron's enterprise customers now that data is on Walrus?
Every dataset they buy now carries a verifiable identifier and an immutable history. Instead of taking Perceptron's word that data is intact, a hedge fund or AI lab can check for themselves that nothing has been altered since collection, and point to that record when compliance asks.
How does Walrus keep data available without a central provider?
Datasets are encoded and spread across a decentralized network of storage nodes, so the data stays available even if up to a third of nodes fail. Retrieval runs at roughly 800ms for sub-1MB files, and the network already holds more than 500TB of unencoded data.
Can other data providers use Walrus the same way?
Yes. Any company selling data into compliance-heavy sectors faces the same question Perceptron did: can you show your product hasn't been tampered with? Walrus works the same way for market data feeds, training corpora, or research datasets, and integration doesn't require rebuilding existing pipelines.
About Walrus
Walrus is the data platform built for the demands of AI. It lets builders, enterprises, and AI agents move data across apps and models, protect it with built-in access rules, and make it verifiable by anyone. As AI takes actions with consequence, the data it runs on has to be provable. Created by the cryptographers and distributed systems engineers behind Sui and Meta's first stablecoin initiative. walrus.xyz
About Perceptron
Perceptron Network is a distributed intelligence infrastructure company powered by 800,000+ user-run nodes across mobile and web. It collects hard-to-source, real-time data at over 90% lower cost than centralized providers - structuring it into high-value signal intelligence products for enterprise, finance, and AI. Contributors earn rewards. Enterprises get premium data.
