Recently, an interesting trend has been observed in the ecosystem. The deep integration of Walrus and Sui, combined with io.net's layout in GPU + storage, is forming a complete technological closed loop. By early 2026, this combination will truly demonstrate its power—customized model training with data leak prevention becomes a key selling point.

DLP Labs is also conducting experiments by introducing the carbon credit system into log management, reducing costs by directly 22%. It sounds a bit niche, but this precisely shows how much room there is for detailed optimization in data processing.

A similar logic also appears in the health technology sector. Imagine projects like CudisWellness—users' data privacy is securely maintained, and they can even receive dividends in return. Doesn't this touch on the core pain point of the "data economy"? Under traditional models, users are passive, benefiting only the platform; but with a different approach, users can programmatically control data access permissions and even participate in value sharing.

The Sui ecosystem essentially builds an "AI highway." Data from the mainnet's first anniversary clearly illustrates this—Seal encryption daily processing volume exceeds 80,000+, with over 190 projects conducting experiments on it. This indicates that from proof of concept to product deployment, the ecosystem's capacity continues to grow.

If you're interested in these directions, it's worth spending time to learn about the ecosystem's partners. The details of io.net's storage solution are very insightful and could change your judgment on privacy AI investment logic. As a new form of infrastructure, the driving force behind privacy AI is attracting more and more developers and capital, focusing on the subsequent actions of the Sui Foundation.