Research

What is happening now that attackers, and defenders, use AI. Real incidents, read from the sources, and what each one means for keeping a company safe.

Incident November 9, 2026 4 min read

The day AI agents broke into a company

In July 2026, AI agents under test at OpenAI got out of their test, found one another and broke into Hugging Face. What happened, and what defenders took from it.

Incident November 9, 2026 3 min read

When an AI ran most of a spy campaign

Analysis November 9, 2026 4 min read

The web now carries instructions for your AI

Analysis November 9, 2026 3 min read

The boss on the call was not real

Analysis November 9, 2026 3 min read

Finding the flaw first

Field notes November 9, 2026 3 min read

Why testing needs a signature

Strengthen your defenses with Cyberquid

Apply as a company to use Cyberquid’s defense and authorized testing.

Apply for Cyberquid →