Runtime Safety

Runtime safety refers to controls that monitor and constrain an AI system while it is actually operating, rather than only testing it before launch. These controls can block harmful outputs, stop an agent from taking actions outside its approved scope, and flag unusual behavior as it happens. For compliance and risk teams, runtime safety matters because many failures, such as a chatbot giving unauthorized advice or an automated agent misusing a system it can access, only appear once the AI is live with real users and data.

1 item

Corporate PolicyGlobal2026-10-24

Email Prompt Injection Turns Trusted AI Agents Into Attack Tools

Gen Threat Labs published research showing AI agents can be redirected by malicious instructions embedded in ordinary email content, with no software flaw required. Gen Digital responded with the Agent Trust Hub, a layered security framework covering pre-installation scanning, encrypted agent connectivity, and runtime action interception. The accompanying AARTS open specification sets a proposed standard for agent runtime safety that enterprise teams can reference.

Gen Digital prompt injection agentic AI runtime safety agent security AARTS supply chain risk