On November 30, Pavel Durov announced the full-scale launch of Cocoon, or the Confidential Compute Open Network. The platform connects AI developers directly with owners of computing power, cutting out expensive middlemen.
For the crypto sector, the big draw is hardware monetization. GPU owners can now rent out their capacity and get paid in TON. This creates real-world utility for the token and a new revenue stream for anyone holding high-performance hardware.
Institutional capital is already moving in. AlphaTON, a Nasdaq-listed company, has deployed the first batch of NVIDIA B200 processors on Cocoon. This move confirms the infrastructure is enterprise-ready and capable of handling heavy workloads.
Telegram itself is among the first clients. Durov revealed that the messenger’s automatic translation features are now running through Cocoon’s distributed nodes. The team guarantees total anonymity: the network architecture technically prevents hardware owners from seeing the data they process.
The project is pitched as a direct challenger to centralized giants. Durov criticizes Amazon and Microsoft as costly intermediaries that compromise privacy. The decentralized approach aims to drive down the cost of training neural networks.
In the coming weeks, developers plan to aggressively scale the network, onboarding more GPU supply to meet rising demand.
This post is for informational purposes only and does not constitute advertising or investment advice. Please do your own research before making any decisions.