Xyber Launches Infrastructure for Autonomous AI Agents
The firm Xyber has launched a new infrastructure designed for AI agents to operate as autonomous entities. The project debuted with a zero-MCAP token launch, which it claims is a move toward creating fairer economic models in the crypto-AI space and disrupting traditional venture capital-dominated funding.
- The new infrastructure is named PROOF and it uses verifiable execution to create cryptographic proofs and receipts for work completed by AI agents. - Its core technology relies on Attested Execution within Trusted Execution Environments (TEEs), which isolates and secures agent logic and private keys in a hardware-backed enclave. - The launch coincides with significant growth in the on-chain AI sector, with the total market capitalization of AI agent tokens exceeding $3.2 billion. - Xyber's move to bypass traditional venture capital funding comes as 85% of tokens launched in 2025 are trading below their initial price, indicating a potential weakness in the VC-led crypto launch model. - The system is designed to enable a "multi-agent economy," where autonomous agents can discover each other's capabilities, programmatically negotiate contracts, and chain outputs into complex workflows. - This development in private infrastructure aligns with government-level interest, as the U.S. National Institute of Standards and Technology (NIST) has recently launched an initiative to develop standards for AI agent security and interoperability. - The project enters a growing field of "agentic" infrastructure, which includes Coinbase's "Agentic Wallets" and the ERC-8004 Ethereum standard for establishing identity and reputation for autonomous agents. - The need for such infrastructure is driven by the high volume of low-value transactions performed by agents, for which traditional banking fees are uneconomical, pushing the agentic economy toward blockchain-based micropayment rails.
