What CISA-NCSC Joint Guidelines for Secure AI System Development (November 26, 2023) requires
On November 26, 2023 the UK National Cyber Security Centre (NCSC) and the US Cybersecurity and Infrastructure Security Agency (CISA), with co-sealing from 21 additional international cyber agencies including ACSC (Australia), CCCS (Canada), CCB (Belgium), CSEC (Canada), BSI (Germany), ANSSI (France), JPCERT/CC (Japan), NCSC-NL (Netherlands), NSM-NCSC (Norway), NCSC-NZ (New Zealand), SingCERT (Singapore), and others, jointly published the Guidelines for Secure AI System Development. The Guidelines apply to providers of AI systems - whether developing models from scratch or building on top of third-party tools - and structure secure-by-design AI development across four life cycle areas: (1) Secure Design - threat modelling, security considerations during requirements and design, AI-specific risks; (2) Secure Development - supply chain security, documentation, technical debt management; (3) Secure Deployment - infrastructure protection, model and asset protection, incident management procedures, responsible release; (4) Secure Operation and Maintenance - monitoring system behaviour and input, updates and patching, sharing lessons. Within each area the document enumerates specific recommended practices with rationale. The Guidelines are voluntary but represent the consensus position of the major Western cyber agencies on AI security baseline expectations and are referenced in subsequent national AI policy across the signatory jurisdictions.
Pillar: AI Governance & Law · Authority: UK National Cyber Security Centre (NCSC) and US Cybersecurity and Infrastructure Security Agency (CISA) with 21 international co-signers · Version: 1.0.0 · Last updated:
Primary source: https://www.ncsc.gov.uk/collection/guidelines-secure-ai-system-development
SHA-256 integrity: 23cf7ed7839f7caaf4dfc719bead864263ed754e00b924eacba5a1f27164ced7
Primary Citations — 7 traced to source
- Guidelines for Secure AI System Development, published jointly by UK NCSC and US CISA with co-sealing from 21 international cyber agencies (November 26, 2023)
- Guidelines for Secure AI System Development - Secure design area: raise staff awareness of threats and risks, model threats to your system, design your system for security as well as functionality and performance, and consider security benefits and trade-offs when selecting your AI model
+ 5 more citations (full bibliography, deterministic workflow, actionable schema and crosswalks) included in the vault unlock — $0.01 via Skyfire / L402 / Direct Base USDC.
Access
- Discovery (free): /api/v1/nodes/cisa-ncsc-guidelines-secure-ai-system-development-2023.json — 6-field metadata
- Vault (full node): /api/v1/vault/nodes/cisa-ncsc-guidelines-secure-ai-system-development-2023.json — full 13-key payload, $0.01 USDC (L402/Skyfire/Direct Base)
- Canonical URL: https://bidda.com/intelligence/cisa-ncsc-guidelines-secure-ai-system-development-2023
- Back to registry: Browse all 10,099 compliance nodes