# Controlling AI agents can feel like corralling toddlers

Canonical: https://brew.new/browse/templates/email/pt1_k97w5azh8hfzdc71g9kb924ean8e5agw

Brand: goteleport.com
Category: newsletter

![Preview of Controlling AI agents can feel like corralling toddlers](https://cdn.brew.new/email-preview-9a8203c4c7ab1b90-tracking_r57zsnzse2gz588a3nryavjx618dj88e-1789006047879.png)

## Email content

Agents as Toddlers: How we get audit & visibility into risky agent action

Agents will follow any path to achieve a goal, and any access they find is a path. Proactively preventing an agent from misbehaving is like corralling toddlers. Ben Arent, Director of Product at Teleport, will show how to detect and prevent misalignment when an agent exhausts the rules rather than breaks them.

When: Thursday, September 17 at 10 AM PT,

What: “Agents as Toddlers: How we get audit and visibility into risky agent action” Webinar

Format: 20 minutes on containing and controlling agents within the boundaries you define, then rapid fire Q&A

Ben will walk through:

Automated summaries of all agent actions, including prompts and responses

Agentic Classifiers that monitor intent and catch agents attempting workarounds

Risk Scoring with MITRE ATT&CK mapping, so a small team can triage agent activity at agent scale

Reserve your spot →

Teleport, 2100 Franklin St, Suite 400, Oakland, CA 94612

If you no longer wish to receive these emails anymore, click on the following link:

UNSUBSCRIBE

TERMS OF SERVICE | PRIVACY POLICY | SECURITY POLICY

[Open and remix this design](https://brew.new/browse/templates/email/pt1_k97w5azh8hfzdc71g9kb924ean8e5agw)

[Browse email designs](https://brew.new/browse/templates)