AI RESEARCH
SplitAgent: A Privacy-Preserving Distributed Architecture for Enterprise-Cloud Agent Collaboration
arXiv CS.AI
•
ArXi:2603.08221v1 Announce Type: cross Enterprise adoption of cloud-based AI agents faces a fundamental privacy dilemma: leveraging powerful cloud models requires sharing sensitive data, while local processing limits capability. Current agent frameworks like MCP and A2A assume complete data sharing, making them unsuitable for enterprise environments with confidential information. We present SplitAgent, a novel distributed architecture that enables privacy-preserving collaboration between enterprise-side privacy agents and cloud-side reasoning agents.