GenAI Guardrails
For All Your Employees
Enable your employees to use AI tools while tackling privacy, IP, and regulatory risks. Detect and control usage, remove sensitive secrets, and transparently log everything in one place.
Control, Not Hope
Safeguard All GenAI Apps
Detect and remove sensitive info from user prompts in ChatGPT, Gemini, Copilot, and more across your entire organization.
Easy & Simple Rollout
Our enterprise-wide GenAI guardrails are installed using a simple browser extension. You’re up in 5 minutes — no proxy, VPN, or complex setup.
Granular Policy Controls
Ready-made, customizable policies allow you to instantly control all GenAI access in one place, including policy violation feedback for users.
Full Usage Observability
Live view of every prompt and policy decision; GenAI usage trends, top violations, signed audits, and access control in one dashboard.
Apply company policies to all GenAI tools used by employees, while blocking sensitive data, recording usage, and fulfilling regulatory compliance obligations.
Remove Secrets
Remove PII/PHI, cloud secrets, and other sensitive info.
Enforce Policies
Define granular user, group, and department policies.
Detect Abuse
Trust scoring enforces good behavior and clean use.
View Detailed Audit Trails
with full prompt, user ID, and other context.
Frequently asked questions.
What GenAI domains does the Iron Book extension protect?
What GenAI domains does the Iron Book extension protect?
Everything comes pre-configured: Out-of-the-box we cover ChatGPT, Gemini, Perplexity, Copilot, and Claude.ai. You can add any other custom domains and use our Policy editor to adjust controls, such as PII/PHI and secrets detection, timezone and region controls, coaching and error messages shown to the users, and more.
Does the Iron Book extension see or store employees’ prompts?
Does the Iron Book extension see or store employees’ prompts?
We securely log and store the prompts in an encrypted immutable audit registry, complete with rich context of every request, such as the user's identity, timezone/region, and more, policy decision, and more. These logs are available for viewing by your security team.
Can we customize the guardrails applied to our employees' prompts?
Can we customize the guardrails applied to our employees' prompts?
Yes. Admins have full centralized control over the guardrails using the Iron Book policy engine. You can add or remove models, regions, MIME types, PII patterns, even budget rules, then push the changes instantly to every browser across your organization.
What happens if Iron Book is down or unreachable?
What happens if Iron Book is down or unreachable?
The extension fails closed after three attempts: Prompts are paused and the user sees a banner. An optional “grace mode” can allow read-only queries during outages.
Is the the deployment process effort-intensive? What's involved?
Is the the deployment process effort-intensive? What's involved?
Simply sign up for the Iron Book portal, then publish the browser extension via Intune, Jamf, or Chrome Enterprise with a pre-seeded API key. No proxy, VPN, DNS config edits, or endpoint reboot required. Template policies and commonly used GenAI apps like ChatGPT and Gemini come pre-configured. Most orgs roll out company-wide in under 30 minutes.
Secure your AI estate, today
Scale your AI innovation securely - let's talk!