{"slug":"uk-principles-safety-security-robustness","id":"AIPT-OBL-0072","permanent_url":"https://aipolicytracker.org/id/AIPT-OBL-0072","title":"Ensure AI systems are safe, secure and robust throughout their lifecycle","short_title":"Safety, security and robustness","category":"risk_management","summary":"Regulators are asked to ensure AI systems function in a robust, secure and safe way, with risks continually identified, assessed and managed. In practice this is enforced through existing safety, security and data-protection law rather than a new duty.\n","practical_action":"Run lifecycle risk assessments and keep security testing evidence that you can show a regulator.\n","is_binding":false,"applies_from":null,"section":"Part 3","source_reference":"Principle 1, Part 3","official_source_url":"https://www.gov.uk/government/publications/ai-regulation-a-pro-innovation-approach","review_status":"verified","confidence_level":"high","last_verified_at":"2026-09-26","reviewed_by":"Bhaskar Bhatt","actors":["provider","deployer"],"sectors":["cross_sector"],"use_cases":["generative_ai","safety_critical"],"applicability":[{"description":"Non-binding expectation for all AI developers and deployers in regulated activities.","actors":["provider","deployer"],"conditions":null}],"evidence_examples":[{"title":"AI risk assessment","description":null,"artifact_type":"report"}],"framework_mappings":[{"framework":"nist_ai_rmf","framework_name":"NIST AI RMF 1.0","reference":"MEASURE 2.5–2.7","note":"Original editorial mapping.","confidence_level":"medium","is_original":true},{"framework":"iso_42001","framework_name":"ISO/IEC 42001:2023","reference":"Clause 6.1 and Annex A risk controls","note":"Original editorial mapping.","confidence_level":"medium","is_original":true}],"controls":[{"slug":"ai-risk-assessment","title":"AI risk assessment and lifecycle risk register","kind":"process","relationship":"satisfies","note":"Lifecycle identification and management of risks.","confidence_level":"medium","url":"https://aipolicytracker.org/controls/ai-risk-assessment"},{"slug":"accuracy-robustness-security-testing","title":"Accuracy, robustness, fairness and security testing","kind":"technical","relationship":"satisfies","note":"Security and robustness evidence to show a regulator.","confidence_level":"medium","url":"https://aipolicytracker.org/controls/accuracy-robustness-security-testing"}],"url":"https://aipolicytracker.org/obligations/uk-principles-safety-security-robustness","policy":{"slug":"uk-ai-regulation-white-paper","title":"A pro-innovation approach to AI regulation (white paper and government response)","status":"guidance","jurisdiction":"uk","url":"https://aipolicytracker.org/policies/uk-ai-regulation-white-paper"}}