Information Security AI Architect

Overview

The Information Security Architect is responsible for designing secure, compliant, and scalable architectures that enable the organization to safely adopt Microsoft 365 Copilot, large language models (LLMs), AI agents, and intelligent automation. This role blends deep expertise in Microsoft Purview, cloud security, responsible AI, and enterprise productivity tooling.

A core mandate includes implementing strong data ring‑fencing controls to ensure sensitive information remains within trusted network boundaries, while also driving measurable business value from Copilot in Microsoft 365 (M365) through secure enablement, workflow optimization, and governed adoption while limiting access to other AI resources from the company network not approved by the Information Security Office.

This individual will lead the company’s M365 CoPilot Purview implementation setting many of the foundational pre‑requisites required for a successful and secure AI ecosphere.

• Design secure architectures for LLMs, Microsoft 365 Copilot, generative AI systems, RAG pipelines, and autonomous AI agents.
• Define guardrails for Copilot and AI agents, including identity boundaries, access controls, and operational constraints.
• Establish governance for prompt security, model input/output filtering, and safe‑use patterns across the enterprise.
• Architect secure integration patterns ensuring all AI systems operate strictly within approved network zones and data boundaries.

• Implement ring‑fencing strategies to prevent sensitive data from leaving the company’s network or being accessed by unauthorized AI systems.
• Design and enforce network segmentation, private endpoints, and isolation zones for AI workloads.
• Ensure Copilot, LLMs, vector stores, and AI agents only access data through approved, monitored, and policy‑controlled channels.
• Build automated controls that block data exfiltration, including DLP, egress filtering, and Purview‑driven policy enforcement.

• Lead secure enterprise adoption of Microsoft 365 Copilot, ensuring alignment with data governance and security policies.
• Identify high‑value Copilot use cases that drive measurable productivity, efficiency, and decision‑making improvements.
• Develop Copilot frameworks, usage patterns, and success metrics demonstrating
• Ensure Copilot interactions respect sensitivity labels, retention policies, and Purview‑driven governance.

• Architect secure AI agent frameworks that automate business processes, knowledge retrieval, and decision support.
• Evaluate and implement AI‑powered copilots, workflow automation, and intelligent assistants across departments.
• Develop standards for human‑in‑the‑loop oversight, escalation paths, and monitoring of autonomous agent behavior.

• Lead the enterprise strategy for Microsoft Purview, ensuring data classification, lineage, and sensitivity labeling support AI and Copilot use cases.
• Implement automated governance workflows that feed LLMs, Copilot, and AI agents only the data they are authorized to access.

• Define and enforce security baselines for AI workloads, including encryption, identity, network segmentation, and model access.
• Ensure compliance with emerging AI regulations, including transparency, auditability, and model risk management requirements.
• Conduct threat modeling for Copilot, LLMs, and AI agents, including prompt injection, data leakage, model manipulation, and supply‑chain risks.

• Serve as the enterprise SME for secure AI and Copilot adoption, guiding engineering, data science, legal, and compliance teams.
• Translate complex AI security concepts into actionable guidance for business leaders and technical teams.
• Lead architecture reviews for AI‑enabled applications, copilots, and agent‑based automation initiatives.

• Implement monitoring for Copilot usage, LLM interactions, agent actions, and data access…