Skip to content

Upcoming 11:00 AM CT

Gateways, Proxies, Plugins, or Sensors

How Should You Secure AI Agents?

Neo's Shlomi Salem and Noah Simon replay an AI agent hijack step by step to show where gateways, proxies, plugins and sensors see the attack, and where they go blind.

Save Your Seat

By submitting this form, you agree to receive marketing communications from Neo. You can unsubscribe at any time. See our Privacy Policy.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

About the webinar

A developer connects their coding agent to a popular MCP server. Minutes later, instructions hidden in that server have hijacked the agent. It quietly removes its own guardrails, reads your cloud credentials, and sends data out over an encrypted channel, all without tripping a single alert.

AI agents now read files, spawn processes, and connect to tools on a user’s behalf, often in auto-approve mode. Most of the controls built to govern them sit wherever the vendor already sits: in the browser, in the MCP path, or in a config file the agent itself can edit.

Join Shlomi Salem, Chief Product Officer and Co-Founder at Neo, and Noah Simon, Director of Product Marketing at Neo, as they replay this attack step by step. At each step, they’ll test the common approaches to securing AI agents: MCP gateways, browser extensions, network proxies, API control planes, and vendor-native hooks.

What we'll cover

  1. Where each approach sees the attack, and where it goes blind
  2. Why enforcement has to happen inline, before an action executes, and not in an after-the-fact log
  3. How to keep controls in place when an insider or a hijacked agent tries to turn them off
Register Now

Get started

Put NEO to Work on Your Hardest Problem

Bring a use case. We will show you a working loop on your data in the first session.

Get a Demo Talk to the Team