Click here to view all sessions in this series.
Your data is democratized. Is your security keeping up?
AI has put the power to build in everyone’s hands. Analysts and data engineers now create semantic views, agents and pipelines every week. Every new asset raises new security questions: who can see this, what does it expose, and what is it costing us?
Join us for a live, instructor-led session where we follow a security investigation from first clue to fix. You’ll watch Cortex Code, Snowflake’s AI coding agent, trace a suspicious data export, uncover hidden access gaps, check whether an AI agent is exposing sensitive data, and put guardrails on AI spend. It explains every finding in plain English along the way. Every fix waits for your approval before it runs.
You don’t need to be a security admin to follow along. If you build on Snowflake, this session is for you. Cortex Code’s prepackaged skills do the heavy lifting, so you can find the gaps and walk into the next conversation with your admin team prepared.
One investigation, four discoveries
1. The Incident
Start with a suspicious data export, and let Cortex Code trace logins, queries and roles back to the source.
2. The Hidden Access Path
Ask “who can see this column, and how?” and uncover the role chain nobody meant to create.
3. The AI Layer
Check whether your semantic views and agents respect the same rules as your tables. Then watch two users ask the same question and get different answers.
4. The AI Bill
Find out who is spending the most on AI, then put budgets and limits in place before the invoice surprises you.
In this virtual session, you’ll learn how to:
- Investigate suspicious activity through conversation, using Cortex Code to connect logins, query history and access patterns
- Understand who can access sensitive data and why, by tracing role hierarchies and grants without writing complex SQL
- Check your account’s security posture, including network access, and how Cortex Code builds on Trust Center findings to prioritize what to fix first
- Secure AI assets end to end, making sure semantic views and Cortex Agents respect masking and RBAC
- Bring AI spend under control with visibility by user, budgets and spending limits
- Fix gaps with approval: review the SQL Cortex Code proposes before anything changes
- Keep the AI itself in check with guardrails that limit what Cortex Code is allowed to do in your session
- See every finding come together in a live command center
Speakers
Kunal Puranik
Senior Solutions Engineer, Snowflake
Register Here