AI Security
Sets out what an organisation needs in place to adopt AI securely. This is a bridge domain: it asks whether the organisation has adopted a security baseline for AI, assesses AI-specific threats, manages the security of its AI supply chain, protects information, tests AI security before release and can respond to AI security incidents. The detailed technical controls are delegated to recognised AI security frameworks, which the baseline in SEC-01 is expected to draw on. Keeping services running when AI fails or gets things wrong is OPM-09.
AI security baseline
The organisation has adopted a recognised AI security baseline, built it into its security standards, and applies it to every use of AI.
AI threat modelling
Each use of AI has its AI-specific threats assessed before it goes live and again when it changes, and the results shape its design and controls.
Security of the AI supply chain
The security of models, data, code and services obtained from others is assessed before use and managed for as long as they are in use.
Preventing information leakage through AI
Sensitive information is protected from exposure through AI tools, prompts, training and outputs, with rules staff understand and technical measures that enforce them.
AI in security incident response
Security incident management covers incidents involving AI, with playbooks, people who have the skills to handle them, and the ability to contain an affected AI service quickly.
Security testing of AI before release
Before release, each AI system is security tested against the threats in its threat model by suitably skilled testers, including checks that it does not disclose personal or sensitive information.