Technical Consultant - Data and AI Security

International Business Machines Corporation; United States; Remote; USD 180,000 - 230,000; Full time

Job summary

IBM is seeking a visionary AI Security Architect to lead the design, assessment, and governance of security frameworks for AI/ML systems across the enterprise. You will be the subject matter expert at the intersection of AI, cybersecurity, and regulatory compliance.

You will collaborate with data science, engineering, legal, compliance, and executive leadership to define how IBM builds trustworthy AI—from threat modeling and policy enforcement to third‑party AI vendor risk and governance.

Responsibilities

• Design and advise Clients on enterprise-wide security reference architectures for AI/ML systems — LLMs, generative AI pipelines, ML inference infrastructure.
• Establish secure-by-design patterns for model training, data pipelines, model registries, APIs, and AI-enabled applications.
• Define security controls for AI infrastructure across cloud, on-prem, and hybrid environments (Azure, AWS, GCP).
• Develop and enforce the organization's AI Governance Framework, including acceptable use policies, model risk management, data ethics, and accountability structures.
• Align AI governance programs with regulations and standards: EU AI Act, NIST AI RMF, ISO/IEC 42001, OWASP LLM Top 10, MITRE ATLAS.
• Define AI risk tiers and corresponding control requirements (high-risk vs. limited-risk AI systems).
• Establish model cards, AI Bill of Materials (AI-BOM), and audit trail requirements for deployed AI systems.
• Partner with Legal, Privacy, and Compliance on regulatory obligations for AI across domains.
• Lead AI-specific threat modeling using MITRE ATLAS, OWASP LLM Top 10, and adversarial ML taxonomies.
• Conduct assessments covering: data poisoning, model inversion, adversarial evasion, prompt injection, training data extraction, supply chain risks, and model theft.
• Assess AI/ML vendor products and third-party AI integrations for security posture and governance maturity.
• Co‑ordinate and run red team exercises — LLM jailbreaking, API abuse, agentic AI manipulation scenarios.
• Produce written assessment reports with risk ratings, findings, and remediation roadmaps for technical and executive audiences.

Job description

We are seeking a visionary AI Security Architect to lead the design, assessment, and governance of security frameworks for AI/ML systems across the enterprise. You will be the subject matter expert at the intersection of artificial intelligence, cybersecurity, and regulatory compliance — ensuring AI systems are built, deployed, and operated securely and responsibly.

You will work across Client's cross-functionally with data science, engineering, legal, compliance, and executive leadership to define how the organization builds trustworthy AI — from threat modeling to policy enforcement to third-party AI vendor risk.

Key Responsibilities

AI Security Architecture

• Design and advise Clients on enterprise-wide security reference architectures for AI/ML systems — LLMs, generative AI pipelines, ML inference infrastructure.
• Establish secure-by-design patterns for model training, data pipelines, model registries, APIs, and AI-enabled applications.
• Define security controls for AI infrastructure across cloud, on-prem, and hybrid environments (Azure, AWS, GCP).

AI Governance & Policy

• Develop and enforce the organization's AI Governance Framework, including acceptable use policies, model risk management, data ethics, and accountability structures.
• Align AI governance programs with regulations and standards: EU AI Act, NIST AI RMF, ISO/IEC 42001, OWASP LLM Top 10, MITRE ATLAS.
• Define AI risk tiers and corresponding control requirements (high-risk vs. limited-risk AI systems).
• Establish model cards, AI Bill of Materials (AI-BOM), and audit trail requirements for deployed AI systems.
• Partner with Legal, Privacy, and Compliance on regulatory obligations for AI across domains.

AI Security Assessments

• Lead AI-specific threat modeling using MITRE ATLAS, OWASP LLM Top 10, and adversarial ML taxonomies.
• Conduct assessments covering: data poisoning, model inversion, adversarial evasion, prompt injection, training data extraction, supply chain risks, and model theft.
• Assess AI/ML vendor products and third-party AI integrations for security posture and governance maturity.
• Co‑ordinate and run red team exercises — LLM jailbreaking, API abuse, agentic AI manipulation scenarios.
• Produce written assessment reports with risk ratings, findings, and remediation roadmaps for technical and executive audiences.

This role can be performed from anywhere in the US.