Tailscale

When an AI Agent Causes the Loss: Shadow AI and the Underwriting Questions It Raises

by

An AI agent can now write code, call tools, move files, and run commands. In most setups, it does all of that with the permissions of the person who launched it. So when an agent causes damage, the first underwriting question is deceptively hard. Who acted? Remy Guercio, Product Lead for Aperture at Tailscale, does … Read more

Zero Trust Meltdown: 83% of Engineers Admit to Bypassing Security in New Tailscale Report

by
A split-screen illustration comparing Zero Trust security with real-world shortcuts. On the left, labeled "Zero Trust," a lock icon connects to circles representing Identity, MFA (Multi-Factor Authentication), and Access. On the right, labeled "Real-World Shortcuts," a computer with a warning triangle is surrounded by icons for passwords, a key symbol, a mobile phone, and a user badge, illustrating insecure practices.

Security Shortcuts Are the Norm, Not the Exception – It’s like locking your car, but leaving the keys in the ignition so you can get going a bit faster next time you head out. Technically locked, practically useless. That’s the state of cybersecurity today, according to Tailscale’s State of Zero Trust 2025 report. The new … Read more