liveBook · Manning

Browser agent

As explained in AI Governance

A browser agent is an AI-enabled browser that can navigate websites and perform tasks such as filling forms across authenticated sessions.

A browser agent can navigate websites, fill forms, and complete tasks across authenticated sessions. Those capabilities combine sensitive data, untrusted page content, and external communication—the Lethal Trifecta. section 4.8.2

Reduce exposure with controls matched to the tasks the agent must perform:

  • Use isolated profiles without password managers, saved credentials, or corporate SaaS sessions. section 4.8.2
  • Use logged-out mode where supported; it reduces capability but removes access to sensitive authenticated-session data. section 4.8.2
  • Use enterprise secure browser solutions to log navigation and agent actions in real time; traditional endpoint monitoring cannot see inside AI browsers. section 4.8.2

How AI Governance covers it

How deep it goes
Explains the risk pattern and operational controls, but does not detail implementation procedures or testing.
Examples
Uses web navigation, form filling, and tasks across authenticated sessions as examples.
What you can do after
Operate/configure browser agents with isolated profiles, logged-out mode, and real-time monitoring of navigation and agent actions.
How it connects
Connects authenticated sessions, sensitive data, untrusted content, and external communication through the Lethal Trifecta.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.