The Nera Blog

Notes from the confidential-AI frontier.

Perspectives on running commercial LLMs over sensitive data, without ever exposing it. Product updates, technical deep dives, and lessons from working with security and compliance teams.

Sovereign AI Has a Definition Problem

Sovereign AI, AI governance, and data residency are being used almost interchangeably right now. They mean different things, protect against different risks, and cost very different amounts. A clear-eyed survey of what's actually being offered today.

Rami Akeela, Ph.D. Read more →

Is MCP Safe for Sensitive Data? What Actually Happens When AI Agents Access Your Records

Most AI integrations protect your data from unauthorized people, not from the AI model itself. Here's the difference between permission-based governance, infrastructure confidentiality, and true model-level isolation, and which industries need which one.

Rami Akeela, Ph.D. Read more →

Can Underwriters Use AI to Analyze Loss Runs and Schedules of Value?

Insurance has spent real money automating loss run extraction. Nobody has built the tool for what happens after: the manual pivot-table analysis underwriters still do by hand, on files full of claimant and policy data.

Rami Akeela, Ph.D. Read more →

The Two Perils of AI Risk: What the Model Says vs. What the Model Sees

AI risk is two different perils wearing one name. Output risk is what the model says. Exposure risk is what the model sees. One is probabilistic. The other is set by architecture, and it can be bounded.

Rami Akeela, Ph.D. Read more →

How Accounting Firms Can Use AI on Client Data Without Exposing It

Accounting firms can use AI on client financials without exposing them. Learn which tasks are safe for public AI, which are not, and how confidential AI lets firms analyze client data the model never sees.

Rami Akeela, Ph.D. Read more →

Why We Built Nera ChatApp to Never See Your Data

Enterprises want to use AI on their most valuable data, but privacy, compliance, and trust concerns often stand in the way. In this article, we explain the design decisions behind Nera ChatApp and how it enables AI on encrypted data.

Ed Chow, CPO, Nera Systems Read more →

How to Run AI on Sensitive Data Without It Leaving Your Control

A practical guide for enterprise data and technology leaders on how to use frontier AI models on sensitive data without your data ever leaving your controlled environment.

Rami Akeela, Ph.D. Read more →

HIPAA and AI: What Healthcare Organizations Can and Cannot Do

HIPAA governs how AI tools can process patient data and most enterprise AI tools don't meet the bar. A practical guide for healthcare technology and compliance leaders on what's allowed, what isn't, and how to use AI on PHI without compliance exposure.

Rami Akeela, Ph.D. Read more →

What Is Confidential AI? The Security Gap Most Enterprises Don't Know They Have

Confidential AI protects enterprise data during computation, not just at rest or in transit. Learn what it is, how it works, and why regulated industries like healthcare and finance need it now.

Rami Akeela, Ph.D. Read more →