Securing the AI agents that fuel your workflows

Register now

* All fields are required

Available live times (choose one)*:

Choose your session

Please choose a session

First name*

Error: Please enter your first name.

Last name*

Error: Please enter your last name.

Business email*

Error: Please enter your email.

Phone number*

This fields is required

Job title*

This fields is required

Company*

This fields is required

Company size*

This fields is required

Country*

This fields is required

AI agents operate on your most sensitive enterprise content — reading files, moving data, sharing documents, and executing workflows at high speed.

Still, most organizations¹ lack the visibility, standards, and governance needed to protect critical business data. In fact, just 39% of organizations report comprehensive visibility across sanctioned and unsanctioned AI use, and only 34% have standards governing how agents access company data. . In this webinar, see how to put deterministic, policy-driven controls in place at the content layer so your security and compliance teams can power AI rollouts with confidence.

^{_{¹ Box Inc., 2026 – The State of AI in the Enterprise}}

Topics we’ll cover:

Defend against prompt injection attacks — Understand how malicious instructions exploit enterprise content and how content-layer controls block threats before they reach your models
Set precise limits on what agents can and cannot do — See how to enforce rules across content access and user interactions (no coding required)
Govern AI agents at scale — Learn how a three-tier permission model lets admins set enterprise defaults, agent authors customize within those bounds, and agent users get clear feedback when actions are blocked or held for review

Speakers

Bakhshi Malhotra

Staff Product Marketing Manager, Box