Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle

What happened

Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle, Google Research announced. Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle October 5, 2026 Eugene Bagdasarian, Research Scientist, and Marco Gruteser, Principal Scientist, Google Research To be useful, AI agents (AI that carries out multi-step tasks rather than answering one question) must understand and be constrained by contextual behavioral norms to ensure they act appropriately. However, realizing this potential requires solving a key challenge: enabling agent capability while ensuring that agents act appropriately.

In this report, you’ll find a breakdown of foundational privacy and security challenges that autonomous agents face today, needing coordinated defenses at the system, model, user, and ecosystem levels. Agentic trade-offs The core challenge of agentic AI is that, for an agent to be useful, it may need to have access to personal data and the ability to take consequential actions across a broad range of contexts. However, this access together with agents’ behavioral flexibility requires meaningfully different approaches from those used in traditional software.

Key facts

• However, realizing this potential — requires: solving a key challenge: enabling agent capability while ensuring that agents act appropriately
• However, this access together with agents’ behavioral flexibility — requires: meaningfully different approaches from those used in traditional software