Security TrustThe hard part of frontier agents is containmentAs agents become better at long-running work and computer use, the product boundary around their capabilities matters as much as the model itself.August 27 2026//1 min read#agents#ai security#cybersecurity#permissionsOpen post→
Agent WorkflowsApproval buttons need contextAn approval step only helps if the human can see the plan, the data, and the rollback story.September 2 2025//2 min read#agents#permissions#uxOpen post→