Anthropic has enabled endpoint detection and response (EDR) monitoring in Claude Cowork to help detect hidden AI agent activity, according to the supplied report dated 3 November 2026.
The development connects Claude Cowork with EDR monitoring, a security approach used to observe activity on endpoints. However, the available information does not identify the EDR products involved or explain how the monitoring operates.
What is confirmed
The supplied material identifies Anthropic and Claude Cowork as the principal organisation and product involved. It describes the purpose of the change as detecting hidden AI agent activity.
No further details are provided about whether the capability is available to all Claude Cowork users, whether it requires configuration by security teams, or what types of activity it can identify. The report also does not state whether monitoring produces alerts, reports or automated responses.
Why the development matters
AI agents can perform actions within digital environments, making visibility relevant to organisations that monitor endpoint activity. EDR monitoring could give security teams a way to look for agent activity that might otherwise remain difficult to identify.
The available report does not establish the scale of the change or describe its effect on users, businesses or security operations. It also does not provide information about privacy controls, retention practices, system requirements or compatibility with existing security platforms.
The next confirmed details would need to come from Anthropic or accompanying technical documentation. Those details would clarify how Claude Cowork's EDR monitoring works, who can use it and what security teams can do with the resulting information.
Conclusion
Anthropic has enabled EDR monitoring in Claude Cowork for detecting hidden AI agent activity, but the technical or rollout details needed to assess the feature more fully is not yet available.
Frequently Asked Questions
Q. What has Anthropic enabled?
Anthropic has enabled EDR monitoring in Claude Cowork.
Q. What is the stated purpose of the monitoring?
Its stated purpose is to detect hidden AI agent activity.
Q. Which product is involved?
The product identified in the supplied information is Claude Cowork.
Q. Does the report name supported EDR products?
No. The supplied information does not name any supported EDR products.
Q. Does the available information explain how the monitoring works?
No. It does not provide technical details about the monitoring process.
