· ai-devtools

Airlock Digital v7.1 releases Agentic Control to block unauthorized AI agent actions

With the launch of Airlock Digital v7.1, the company has introduced Agentic Control, a feature set that brings specialized Agentic Steering for artificial intelligence agents.


This new capability permits users to broaden their existing Deny by Default Application Control policies to encompass this rapidly developing technology. Instead of merely focusing on what individuals are permitted to execute on their devices, Application Control is now evolving to address AI. Airlock Digital now scrutinizes the actions an agent requests while it is running, thereby halting any unauthorized agent conduct before it has a chance to run. This development, which allows Airlock Digital to block unauthorized AI agent actions at the endpoint, was first published on Help Net Security.

Why it matters

This update extends endpoint security beyond traditional employee application control to govern AI agent actions. Developers and security teams should note this as a new layer of runtime control for emerging AI technologies.

Related coverage

Also worth reading

  1. next.js 16.5.0-canary.56 released
  2. Angular 22.2.7 released, fixes hydration warnings, improves forms date formatting
  3. next.js 16.5.0-canary.54 released
  4. Satya sorry for SaaS-y SNAFU - look over there, more AI!
  5. Website management platform Optimizely confirms hackers put malicious code into customer sites