Blog

Insights

AI governance, security, and enterprise compliance from the Arbitex team.

1 post
Product7 min read

What We Shipped This Quarter: A Q2 2026 Product Update

A roundup of what Arbitex delivered this quarter — DLP enforcement across every channel, enterprise governance that scales with the org, and air-gapped Outpost with full detection parity.

1 post
Security4 min read

Announcing Arbitex's Coordinated Vulnerability Disclosure Program

We're opening a front door for security researchers — with written safe-harbor, PGP-encrypted reporting, response SLAs, and public credit. Here's how to reach us and what to expect.

1 post
Security6 min read

MCP Security: Governing AI Tool Calls Before They Execute

The Model Context Protocol gives AI agents access to databases, APIs, and file systems. Here is how to govern what those agents can actually do.

2 posts
Engineering5 min read

Building Arbitex: Engineering Principles

How we build enterprise security software — testing discipline, security-first design, container hardening, and DLP accuracy methodology.

Security7 min read

Why AI Security Needs a New Approach

AI tools are standard equipment for enterprise teams. Traditional security wasn't designed for this. Here's what a purpose-built approach looks like.

1 post
Engineering6 min read

Rust-Powered Core Loops for Performance and Code Safety

Why we built a compiled native extension for the DLP pipeline's performance-critical inner loops — delivering speed and memory safety without a full rewrite.

1 post
Engineering7 min read

AI Gateway Security Architecture: Four Core Patterns

How Arbitex secures AI traffic with AES encryption, HMAC-chained audit logs, multi-layer content inspection, and zero-trust identity.

3 posts
Security7 min read

BYOK Encryption: Why Your AI Audit Logs Should Use Your Keys

How envelope encryption with a KMS abstraction layer makes bring-your-own-key work for AI gateway audit logs in regulated industries.

Engineering7 min read

How We Optimized the DLP Pattern Matching Engine

A compiled single-pass pattern matching engine processes all DLP patterns simultaneously — with tier optimization and smarter contextual validation.

Security7 min read

Email DLP: Extending the AI Gateway's Protection to Enterprise Email

The same 3-tier DLP pipeline that inspects AI prompts now scans email — with TLS, attachment extraction, compliance templates, and quarantine management.