Agent identity & access
Treat an agent as an actor with an owner, a purpose, and permissions that can be inspected and revoked. Follow what it can do across connected systems.
Who owns the agent and the access behind it?
Read this topic →Cybersecurity for accountable leaders
Practical security reading for the decisions that cannot wait: agent access, AI defense, and incident response. Clear ownership before the pressure arrives.
Choose your next question
Treat an agent as an actor with an owner, a purpose, and permissions that can be inspected and revoked. Follow what it can do across connected systems.
Who owns the agent and the access behind it?
Read this topic →Connect model behavior to application controls. Focus on the data, tools, and business actions that turn a bad answer into a consequential incident.
What could a manipulated answer actually cause?
Read this topic →Prepare decision-makers as well as responders. Establish authority, credible communications, and recovery criteria before an incident compresses the timeline.
Who can make the next critical decision?
Read this topic →Latest thinking & refreshed archive
New analysis alongside refreshed original articles, ordered by publication or original issue date. Each article clearly identifies its edition.
October guidance from U.S. agencies and a Singapore tabletop exercise put agent control and incident readiness in focus. Here is a bounded drill for testing who can stop an agent and establish what it changed.
A connected agent needs an accountable owner and a bounded identity. Start with what it can reach, then test what it is allowed to do.
Treat the software delivery pipeline as privileged infrastructure, with explicit trust boundaries from source changes to production.
Voice impersonation changes what a familiar call can prove. Build verification into urgent payments, access changes, and sensitive requests.
MFA protects sign-in, while application consent creates another access decision. Review both before trusting connected tools.
Defend the application around the model: the inputs it trusts, the data it reaches, and the actions it can take.
Give employees a useful approved route for AI work, with clear boundaries for data, integrations, and consequential decisions.
Improve patching speed by removing ownership and validation delays while keeping priority tied to actual exposure.
Seven practical decisions turn recovery targets into a service the business can actually resume.
A response plan is a decision system. Rehearse authority, communication, and recovery, not just the technical investigation.
Revisit seven ransomware assumptions with a practical recovery checklist, updated for current leadership discussions.
Turn a broad security score into a specific business scenario, an evidence-based estimate, and a decision someone owns.
Translate recovery ambitions into business priorities, decision authority, tested dependencies, and measurable evidence.
A practical reading room
Explore the decisions behind the technology. Explore articles and discussion guides for readers interested in the decisions behind the technology.
About this publication →Keep exploring
Read the latest analysis, explore a topic in more depth, or find speakers with relevant public work.
May we use Microsoft Clarity cookies, heatmaps, and session recordings to understand how readers use this site? Analytics stays off unless you allow it. Advertising storage is disabled.
Privacy details. You can change your choice using Analytics preferences in the footer.