Government leaders see AI-based agents as powerful collaborators that can work alongside humans to improve efficiency and advance missions. However, agentic AI introduces new security and identity challenges at scale that traditional identity and access management solutions, built primarily for human users, are not designed to address or govern.
Identity and access management solutions can address these potential threats from AI agents with AI-enabled solutions. Traditional IAM products do not work as effectively to enable and manage AI agents, given their speed and autonomy. Even products designed to manage the identities of software bots and devices struggle to keep up. “To meet these challenges, agencies need to rethink their approach to identity management”, said Kelvin Brewer, Public Sector Field CTO at Ping Identity.
“In a lot of ways, we will treat agents a lot like we do normal users,” said Brewer. “But there are different protocols and different factors that have to be considered, and differences in how we manage the interactions between users, non-person entities and AI agents.”
In this video interview, Brewer discusses how agencies can deploy AI agents safely and at scale. Topics addressed include:
- Key risks introduced by AI agents
- How identity for AI differs from traditional identity solutions
- How AI can help detect malicious, unauthorized agents