Ahmadbek

About

I’m an independent researcher from Tajikistan working with a seven-person team on AgentTrust, a project studying security and trust boundaries in AI agents. We examine whether untrusted external content can lead an AI agent to take actions the user did not authorize. Our experiments use local tool stubs, so they do not affect real accounts, databases, or email systems. We’re interested in responsible, reproducible research on prompt injection and practical ways to make AI agents safer. I’m open to research collaboration.

Research keywords

Other Artificial Intelligence researchers

Other researchers at Independent Researcher