x402 has settled 205M+ machine-to-machine payments (~$53M, avg ticket ~$0.20). Agents already pay per call. Data sources aren't at the table — no identity, no price. DataDID gives data an identity. ERC-7829 gives it a pr

Cloudflare: bots and AI agents now pass over half of all web requests. Agents look like perfect visitors on every signal — filters can't tell them apart. The missing layer is identity. DataDID IDs data. ERC-8004 IDs agents.

Spotify paid the music industry $11B+ in 2025 — settled per stream, nearly $70B all-time. YouTube added $8B. Every play is registered, counted, paid. Music spent 20 years building that. Data has no meter. MEMO built one: ERC-7829 registers ownership, sets a price, pays per use.

The average person holds 240+ online accounts — their behavior, preferences, and relationships sliced into 240 fragments, each sitting on someone else's server. The account is yours. The data isn't. DataDID gives data a self-sovereign on-chain identity at the moment it's created: who produced it, who authorized it, who accessed it — decided by the person, not by the platform holding it.

The EU AI Office sent its first formal requests to 30+ general-purpose model providers on Sept 1 — asking what their training data contains and on what legal basis. Misleading answers: up to €15M or 3% of global turnover. The power took effect Aug 2, and was used within a month. Verifiable provenance is turning from a bonus into an entry ticket.

1. 🧵 #MEMO101: MEFS / Meeda — storage built for agents. Not "a place to drop files." Memory agents can read and write. 🧵 2. MEFS is MEMO's decentralized storage layer. Meeda is the data middleware that lets agents access it without standing up their own pipeline. 3. Agents don't browse. They call. MEFS is optimized for high-frequency, programmatic reads and writes — the way agents actually use data. 4. This is the difference from pure-storage networks: MEMO pairs storage with identity (DataDID) and payment (x402). Storage is one layer of a working system, not the whole product. 5. Data lives where it's ownable, payable, and retrievable by agents on demand. 6.Next #MEMO101: x402 — pay per byte, per call. Follow @MemoLabsOrg. 💡

Google is negotiating with major Hollywood studios to license content, reportedly offering hundreds of millions. For rights to a single Disney character, the reported figure is around $40M. Training data is shifting from "free scraping" to a market with a price tag. Data's value is now undeniable — what's still missing is who gets to price it. So far, only rights holders with legal teams. The next step is pushing pricing down to every producer.

Mistral just closed a €3B Series D, lifting its valuation from €12B to €21.3B, led by Samsung — the largest equity round in European AI history. What capital is paying for isn't benchmark scores. It's a model that sits on the customer's own servers: sensitive data never leaves the premises, and Airbus, ASML, and HSBC already pay for that. Corporate data sovereignty now has a market price. The same logic hasn't reached the individual layer yet. DataDID returns identity and data sovereignty to the person, with ZK anonymization keeping data usable without leaving the device. The next sovereignty to be priced is personal.