The more control AI agents gain on-chain, the more important it becomes to know who stands behind their actions. These agents can already execute transactions, manage assets, and make decisions across digital systems with less direct human involvement. But when something goes wrong, responsibility should not become unclear. This is the gap @Concordium is addressing at the protocol level. Through its identity-first infrastructure, an AI agent can be connected to verifiable accountability without forcing the human or organisation behind it to expose sensitive personal information. Giving AI agents more access should not make their actions anonymous or responsibility harder to trace. If they are going to manage real value across crypto, users need to know who approved their actions and who answers when something goes wrong. AI may reduce direct human involvement, but accountability should always remain part of the system. #Concordium $CCD