HACCP + Natasha Law + FSA + ISO 22000. One substrate.

HACCP itself is technology-neutral: the 7 principles (hazard analysis, CCP identification, critical limits, monitoring, corrective actions, verification, documentation) apply whether monitoring is manual or AI-driven. But if AI performs CCP monitoring — e.g. computer vision for cook temperature or fill level — you must validate it as a monitoring procedure under Principle 4 and keep verification records under Principle 6. MEOK ships the HACCP AI monitoring + verification evidence as a signed pack.

Natasha's Law (in force since October 2021) requires Pre-Packed for Direct Sale food to carry a full ingredient list with the 14 regulated allergens emphasised (e.g. bold). If you use AI to generate or check labels, the AI must capture every ingredient and never drop an allergen flag — a missed allergen is a criminal offence and a fatal safety risk. Our label-compliance MCP cross-checks generated labels against the 14 allergens and the recipe source of truth.

The FSA inspects against the FHRS using 6 categories and issues a 0–5 rating. AI for hygiene-violation detection (e.g. CCTV monitoring of handwashing or cross-contamination) is a supporting control, not a substitute for the statutory inspection. Keep AI outputs as auditable evidence the inspector can review, and ensure corrective actions are logged — that is what improves the rating.

Allergen preferences and dietary restrictions can reveal health information, which is special-category data under GDPR Article 9. If your AI processes customer allergen/dietary data (e.g. personalised menus, allergy pre-orders) you need an Article 9 condition (usually explicit consent) plus the lawful basis under Article 6, data minimisation, and retention limits. ISO 22000 communication requirements and GDPR must be reconciled — our GDPR MCP handles the special-category mapping.