AI Security Architect

About the role

The AI Security Architect will own the security posture of AI tooling across the Marco Group, covering platforms such as Microsoft Copilot and Anthropic Claude, AI‑assisted development tools and third‑party AI usage. Reporting to the Group Head of Information Security, the role defines safe, rapid AI adoption in line with regulatory expectations.

Key responsibilities

• Own security configuration and hardening of Claude and Microsoft Copilot, including connectors, agentic features and data‑access scopes.
• Define acceptable‑use policies, permission boundaries and DLP controls for generative AI tools.
• Design and run an AI risk assessment framework covering data classification, model access and output integrity for every new AI use case.
• Maintain a central AI tooling register with risk ratings and compensating controls.
• Partner with engineering teams to embed security guardrails in AI‑assisted coding tools and define secure‑by‑design patterns for internal AI solutions.
• Assess AI capabilities in third‑party/vendor products, contribute AI‑specific clauses to contracts and monitor shadow AI usage.
• Report AI risk posture to senior leadership and support regulatory alignment (DORA, UK GDPR, sector guidance).

Required profile

• Strong background in information security architecture, preferably with cloud/SaaS experience.
• Working knowledge of large language models, generative AI architectures and associated threat models.
• Familiarity with regulatory frameworks such as DORA, UK GDPR, ISO 27001 and NIST CSF.
• Proven experience running structured risk assessments and translating findings into technical controls.
• Excellent stakeholder management, able to engage both technical teams and senior leadership.

Required skills

• Microsoft Copilot
• Anthropic Claude
• Large Language Models (LLM) and generative AI architectures
• Cloud and SaaS security
• Vendor and AI risk assessment frameworks
• AI governance frameworks (ISO/IEC 42001, NIST AI RMF)