DATA Network is migrating to Sui, and the Sui Foundation is putting $30 million behind the move. The investment gives the foundation the largest institutional stake in the network, which will now run as a protocol natively on Sui. Its independent chain is being retired.
The migration brings Trace, DATA's rights-cleared AI training data registry, onto Sui's infrastructure. Trace has logged more than 250 million contributions since launching in June and now processes over 3 million receipts per day on average.
Walrus becomes the storage layer for DATA apps
Walrus, a data platform with integrity proofs anchored on Sui, is now the default storage layer for applications operated by DATA Foundation. The pairing addresses a specific problem: AI datasets are enormous, and licensing them requires more than a metadata receipt. Walrus stores large datasets efficiently and ensures only licensed buyers can read them.
With Walrus, Trace can hold an encrypted, immutable copy of the contributed file itself as a reference guarantee. That means the record hash can be regenerated from the data itself rather than from metadata an application supplied. It's a meaningful distinction for anyone who has tried to verify training data provenance after the fact.
The $30 million check and a new advisor
The Sui Foundation's $30 million investment is the largest institutional stake in DATA Network to date. Separately, SY Lee — chairman of Poseidon and CEO of PIP Labs — is joining the Sui Foundation as a strategic advisor.
For existing $DATA holders, the token continues as the token of the DATA protocol on Sui, carried over one-to-one. Holder economics and governance rights are at least equivalent to today's terms, and locked and vesting positions carry over unchanged.
Who's already writing to Trace
Kled, described as the largest opt-in human data marketplace, is onboarding more than 1.5 billion records to Trace. Numo and Oto are also writing to the registry; both already license data to frontier AI labs and robotics companies.
The demand side is being pushed by regulation as much as by research needs. The EU AI Act and ongoing litigation over training data have turned provenance from a nice-to-have into a procurement requirement for AI labs. Buyers need to know where a dataset came from and that the rights are cleared.
Why Sui's architecture matters here
Sui represents every asset as a programmable object. Datasets and their licenses can be managed onchain like tokens, which means a license isn't just a PDF in a folder somewhere — it's a thing with rules attached. Independent transactions execute in parallel and settle in under a second, a throughput characteristic that matters when you're moving AI-scale data.
The migration doesn't happen overnight. DATA Network's independent chain will be retired as the protocol moves onto Sui, and the foundation hasn't published a specific decommissioning date. For now, the pieces are in place: the capital, the storage layer, the applications, and the regulatory pressure. What remains is execution — and watching whether 3 million daily receipts becomes 30 million.
