Jamf Blog

Isolation and authorization limit the damage when an AI agent fails, but they can't stop it from misreading inputs. This final part covers how to treat retrieved content as untrusted, log agent activity independently and plan for recovery.

Part 2 of Jamf's agentic security series moves from containing code execution to constraining enterprise authority. Learn about scoping agent identities and credentials, governing MCP servers and tools, and enforcing authorization outside the model.

Amazon's same-day delivery works because its AI is governed. This AWS session at JNUC 2026 shows how to bring that same control to enterprise AI on Amazon Bedrock.

AI agents are powerful tools, but they introduce unique risks to your organization. Learn how AI platforms within your company cause trouble, from both attackers and legitimate use, and how AI governance helps defend against these threats.

AI tools can accelerate data exposure, compromise software supply chains and trigger risky agentic behavior. Part two of a two-part series.

Black Hat USA 2026 highlighted how cybersecurity has fundamentally changed. Here's a peek into what the Jamf team was up to, what we learned, and what we showed at the conference.

AI use in your organization can be a silent threat multiplier. Here's how to ensure your organization's data remains secure. Part one of a two-part series.

The EU AI Act's high-risk deadlines just moved. Here's what stayed, what changed and what to do now.