Securing Non-Human Identities in the Age of AI Agents | CSA Summit 2025 at RSAC
Released: 04/29/2025
Updated: 11/17/2025
Today, there are 45 non-human identities (NHIs) for every human identity—a figure expected to rise with the increasing adoption of AI agents. Gartner predicts that by 2028, at least 15% of day-to-day work decisions will be made autonomously through agentic AI. These AI agents utilize NHIs—such as service accounts, API keys, and secrets—to access and operate on organizations' most sensitive systems and data. Without proper security measures, these identities become prime targets for attackers, as seen in the past year in high-profile attacks against the U.S. Treasury, Okta, Cloudflare, and Snowflake.
This session explored strategies for enterprises to mitigate these risks by implementing effective NHI security measures. Topics include best practices for visibility, governance, and continuous monitoring to prevent identity-based attacks. Albert Attias, Senior Director of Security Architecture at Workday, shared insights from two years of managing an NHI security program. Astrix’s CEO, Alon Jackson, discussed approaches to future-proof identity security in an AI-driven landscape.
Presented by:
This session explored strategies for enterprises to mitigate these risks by implementing effective NHI security measures. Topics include best practices for visibility, governance, and continuous monitoring to prevent identity-based attacks. Albert Attias, Senior Director of Security Architecture at Workday, shared insights from two years of managing an NHI security program. Astrix’s CEO, Alon Jackson, discussed approaches to future-proof identity security in an AI-driven landscape.
Presented by:
- Alon Jackson, CEO, Astrix Security
- Albert Attias, Senior Director of Security Architecture, Workday
