We research how emerging AI protocols can be exploited for data theft. Our work for BlackFog examines how the Model Context Protocol (MCP) could be weaponised as a covert channel for data exfiltration, an attack vector that is only now beginning to receive attention as agentic AI adoption accelerates.
“Daniel is extremely knowledgeable about the cybersecurity and threat landscape. His research is thorough, and his writing is publishing-ready.”