Story Protocol just became DATA Foundation. The AI receipt idea is clean, but adoption is the catch.
DATA's strongest move is keeping private data private while making consent checkable. The weak point is whether AI labs will accept a shared registry they don't control.
CryptoVibe Desk · ai · data-provenance · story-protocol
- →Story Protocol rebranded as DATA Foundation and shifted from broad IP infrastructure to AI training data provenance.
- →The design publishes cryptographic receipts for data contributions, not the data itself, which is the right privacy tradeoff.
- →By the end of 2026, the real test is whether a major AI lab uses the registry for live dataset verification.
- provenance → Provenance means the history of where data came from, who approved it, and what terms apply.
- cryptographic receipt → A cryptographic receipt is a public proof that something happened without showing the private data behind it.
- content hash → A content hash is a short fingerprint that changes if the original file or record changes.
- onchain registry → An onchain registry is a public blockchain record that apps can check instead of trusting one company's database.
DATA Foundation wants AI provenance to look like a receipt.
Story Protocol just rebranded as DATA Foundation. The a16z-backed Palo Alto startup is moving from general IP infrastructure into AI training data provenance. The company will run DATA Network, an onchain registry for dataset origins, licensing, and consent history. It will also launch Trace, a public audit platform for individual data contributions.
The design bet is simple. Publish the proof, not the private data. Trace records content hashes, consent terms, licensing, payment proof, and timestamps. The underlying data stays inside the marketplace. Access still needs a licensed transaction.
That is the clearest build in this category so far. At the contract level, DATA is treating provenance like replay protection. You don't need the whole payload in public. You need a receipt that can be checked later by someone outside the original deal.
CoinDesk reported on June 25 that DATA has $140 million in cumulative venture funding led by a16z crypto. The same report said Kled founder Avi Patel joins as chief data officer. Andrea Muttoni becomes CEO. CoinDesk also reported that a Kled AI integration has registered 1.1 billion user-contributed records. That number is company-reported, so don't treat it like independently verified chain data.
The reported $2.4 billion valuation is even softer. Gate reported it, and Avi Patel declined to confirm it. If you're reading this as a protocol story, the valuation is not the only number that matters. The 1.1 billion record count is the claim that has to survive inspection.
DATA also operates Poseidon, which cleans and scores human data through Numo. Numo pays contributors stablecoins in real time, with fiat rails running in parallel. That matters because payout timing should not depend on one buyer closing one deal. The engineers know this problem: backpressure in payments breaks contributor supply.
The controversy is still there. Story Protocol drew heat in February 2026 for delaying a token unlock. Co-founder Sy Lee said the chain needed more time to build usage. That makes this pivot harder to grade on announcement alone.
The strong version of DATA is an open receipt layer for AI training data. The weak version is a marketplace with a blockchain audit log attached. And that's the catch: if OpenAI, Anthropic, Google, or Meta build private provenance systems instead, third-party checks stay outside the room.
DATA Foundation's receipt-first design is the right call. But the moat is narrower than the announcement implies because OpenAI, Anthropic, and Google can make shared verification ornamental by staying private.
By the end of 2026, watch whether at least one major AI lab publishes Trace-compatible receipts for licensed datasets. If not, DATA is still a marketplace wrapper.
Primary links and supporting reads used by the desk for this story.
Forward this.
