Noma Security

[◕_◕]📤 your agent before trust boundaries
ᕦ[•̀ᴗ•́]ᕤ your agent after

Where do you start securing AI workflows? Our CISO, Diana Kelley, maps 3 trust boundaries ⤵

Actions, retrieval, and memory

Read the blog: noma.security/blog/thats-a-g…

Things that don’t exist:

❌ A fully automated AI org
❌ A smooth process
❌ The pressure you think you’re under

Dr. Cari Miller’s advice for AI buyers, on a new episode of The Blast Radius with Noma CISO Diana Kelley: trust your gut.

Full episode: youtu.be/aI1UrZBDAhg

Our CISO keeps hearing the same gaps from other CISOs as AI agents scale.

They add up to 5 security shifts, and most point to one idea: the controls have to live outside the model.

From our CISO to yours 💌👇

Teams build apps and agents on @Base44 every week.

Now security teams can see all of it in Noma 🫧

Noma is live on the Base44 Compliance API, so Base44 Enterprise customers can bring every app, AI agent, integration, and member in their workspace into Noma, next to the rest of their AI security.

Build away 🏗️

We’ve increasingly seen agents hold up against single attacks and then fail when those same techniques are combined.

Here, we’re sharing how our next-generation red teaming engine tests for that, backed by Noma Labs research.

Full update: noma.security/blog/next-gene…