AI risk mapped to authority.
Each brief translates a major AI risk into one question: what authority should be required before AI is allowed to act?
View Authority Risk IndexDeepfakes and Synthetic Media
Deepfakes become an execution risk when synthetic content triggers payments, approvals, identity checks, or business workflows.
AI Privacy and Synthetic Identity
AI privacy risk is not only about access to information. It is about whether AI can use, share, or act on that information.
AI Companions and Child Safety
AI systems interacting with minors require stronger escalation rules, authority boundaries, and human intervention requirements.
AI Market Manipulation
Financial actions are irreversible, high-impact decisions that should require pre-execution authorization.
Bias Becoming Action
The key question is not only whether a model is biased, but whether a biased recommendation is allowed to become an action.
AI Job Replacement and Task Authority
As AI moves from assisting work to performing work, companies need policies defining which tasks AI may perform autonomously.
Loss of Control Over Digital Systems
Government AI, enterprise operations, and AI-managed systems need fail-closed execution governance.
Biosecurity and AI Execution Risk
Dual-use scientific systems require strict human approval before AI-assisted biological or laboratory execution.
National Security AI Risk
National security AI risk is increasingly about permission: whether AI should be permitted to act.
Briefing table
Scan ModeUse these briefs as the publishing engine.
Each risk brief should be refreshed when new examples appear, while the guide pages remain evergreen search assets for approval workflows, access control, audit trails, runtime governance, and tool call security.