Algorand's AC2 Keeps AI Agents Off Users' Private Keys
The Algorand Foundation has introduced AC2 (Agentic Communication and Control Protocol), an open protocol designed to let AI agents request sensitive actions without holding users' private keys or API credentials.
AC2 is meant to work like a doorbell, where the agent asks for access and the user approves on their own device, keeping credentials secure. The protocol combines three open standards: DIDComm v2.0 for message formatting, WebAuthn/FIDO2 for phishing-resistant authentication, and WebRTC DataChannels for peer-to-peer transport.
The launch of AC2 aims to give agents enough authority to be useful without giving them the authority to act against a user's interests. Developers building agentic commerce and automation tools can now opt for human-in-the-loop control that emphasizes credential isolation and verifiable intent.