AI Governance Principal

Novelis•Atlanta, GA

2d•Onsite

About The Position

Novelis is one of the world leaders in aluminum recycling and rolling and a leading sustainable aluminum solutions provider. Driven by our purpose of shaping a sustainable world together, we work alongside our customers to provide innovative solutions to the aerospace, automotive, beverage packaging and specialty markets. Headquartered in Atlanta, Georgia, Novelis has approximately 13,000 employees in 32 operating facilities on 4 continents. Novelis is seeking a AI Governance Principal to own the operational governance gates for AI systems across the enterprise, ensuring AI solutions meet established quality, performance, and lifecycle standards prior to deployment and throughout production. This role is responsible for overseeing AI-specific operational risks—including model drift, hallucinations, bias management, explainability implementation, and autonomous or emergent system behaviors—and for working closely with AI delivery teams to ensure these risks are effectively managed. Reporting directly to the VP of Data, Analytics & AI and based in Atlanta, GA, this role is organizationally independent from the AI delivery function to maintain governance objectivity, consistent with the governance independence principles outlined in the NIST AI RMF Playbook. This role carries a program-management and regulatory compliance orientation. Given Novelis’s multinational footprint and exposure to aerospace clients, the AI governance function must be built to navigate EU AI Act requirements, NIST AI RMF standards, ISO/IEC 42001 obligations, cross-border data protection regulations, TISAX certification requirements, and sector-specific regulatory risk. This role reinforces the enterprise Data & AI Governance framework without duplicating data governance controls. The AI Governance Principal maintains clear separation from cybersecurity AI governance - which owns security threat models, penetration testing, and SOC integration. This role is accountable for demonstrating ongoing compliance with all required governance standards, including cybersecurity governance, and holds the authority to require remediation or suspend production deployment when governance standards are not met.

Requirements

Bachelor’s degree in Computer Science, Data Science, Information Systems, Law, or a related field.
Minimum of 7 years of experience in AI governance, AI ethics, responsible AI, or AI risk management, with at least 3 years directly defining and operationalizing AI-specific governance frameworks.
Demonstrated experience defining AI model audit protocols, explainability standards, bias testing procedures, or AI risk assessment methodologies.
Working knowledge of AI/ML system lifecycles to serve as a credible governance authority with AI engineering teams.
Familiarity with the AI regulatory landscape including the EU AI Act, NIST AI RMF, ISO/IEC 42001, or equivalent.
Strong communication skills with the ability to translate regulatory and technical AI governance requirements into enforceable policies and represent the function in executive forums.

Nice To Haves

Master’s degree or advanced certification in AI ethics, responsible AI, data science or law.
Juris Doctorate.
Certifications such as CAIP, ISO/IEC 42001 Lead Implementer, ISACA AI Fundamentals, or equivalent.
Experience in manufacturing, industrial, or sustainability-focused organizations.
Experience establishing AI governance programs from the ground up in organizations deploying AI at scale across multiple business functions.
Experience governing enterprise generative AI tool adoption (e.g., Microsoft Copilot) including acceptable use policy development and output governance.
Familiarity with TISAX certification requirements and cyber liability insurance considerations.
Experience with cross-border AI deployment governance in multinational organizations.

Responsibilities

Establish and operate pre-deployment governance gates—bias and fairness testing, explainability validation, safety guardrail verification, and documentation completeness—and serve as the governance approval authority for AI production readiness.
Enforce ongoing production governance including drift detection thresholds, retraining approval criteria, and periodic model reviews.
Maintain and publish model card templates aligned to EU AI Act requirements, including tier classification worksheets and validation and pre-deployment checklists.
Maintain authority to require remediation or suspend production deployment when governance standards, including cybersecurity governance, are not met.
Operate the AI model inventory and registry within the enterprise governance platform, Informatica Cloud Data Governance and Catalog (CDGC), ensuring all production AI models are cataloged, classified, and traceable.
Own the AI use case intake process, including use case templates, architectural pattern validation, and model onboarding workflows.
Ensure every new AI initiative undergoes comprehensive evaluation across model selection, security and data risk review, data quality assessment, and governance compliance before proceeding to development.
Own EU AI Act conformity assessment templates and geographic deployment scope tracking for all production AI systems.
Align AI governance practices with NIST AI RMF, ISO/IEC 42001, and applicable cross-border AI deployment regulations.
Maintain and operate the enterprise AI risk register, ensuring all identified AI risks are documented, assessed, mitigated, and auditable.
Conduct vendor AI due diligence for third-party AI components and maintain the vendor AI due diligence checklist.
Operate the embedded AI review process for AI capabilities within SaaS platforms, ensuring governance coverage extends to procured AI features.
Oversee model validation, accuracy, robustness, and drift detection standards for all production AI models.
Define and enforce quality and reliability standards for agentic AI behavior, including autonomous decision boundaries and exception handling.
Govern enterprise generative AI tool adoption (e.g., Microsoft Copilot), including development and enforcement of acceptable use policies and output governance standards.
Own and maintain the AI-specific incident response playbook, including escalation protocols, root cause analysis, and remediation tracking.
Define and enforce AI safety guardrail standards across all deployed AI systems.
Coordinate with Cybersecurity on AI-related security incidents, maintaining clear escalation and handoff protocols.
Define and enforce agent permission and tool scoping standards for both self-service agents and managed agents.
Validate human-in-the-loop design compliance for all autonomous workflows prior to production deployment.
Establish governance controls for multi-agent workflows, ensuring behavior predictability, auditability, and graceful degradation.
Represent the AI governance function in the AI Steering Committee, executive forums, and cross-functional governance discussions.
Translate regulatory and technical AI governance requirements into enforceable policies understood by business, engineering, and leadership audiences.
Coordinate with Cybersecurity AI Governance to maintain clear, documented boundaries between platform governance and security governance responsibilities.
Coordinate with the Manager, Data Governance to certify that AI training data meets provenance, bias screening, and quality standards before models proceed through governance gates.
Operate the AI model inventory and registry as a governed tenant within CDGC, following platform standards set by the Manager, Data Governance.
Coordinate on configuration changes, access provisioning, and platform upgrade impacts to the AI governance module.
Align AI data access requirements with the data classification, privacy, and entitlement standards enforced by Data Governance, ensuring AI systems access only appropriately classified and governed data.
Coordinate joint audit and regulatory responses where inquiries span both data governance (lineage, access controls, classification) and AI governance (model audit, regulatory compliance, risk register), delivering a unified governance narrative.
Maintain a shared escalation protocol with Data Governance for incidents at the intersection of data quality and AI model performance, ensuring clear ownership: Data Governance owns the data quality incident, AI Governance owns the AI impact assessment and remediation decision.
Ensure the AI governance framework reinforces the enterprise Data & AI Governance framework without duplicating data governance controls.
Contribute to quarterly planning, feature scoping, and sprint execution aligned to the enterprise delivery roadmap and KPI framework.
Build trust, drive alignment, and strengthen adoption of AI governance practices across business, technology, risk, legal, and cybersecurity stakeholders.
Lead through expertise, sound judgment, and governance authority to align stakeholders on AI risk, policy, and control requirements across the enterprise.
Build strong working relationships with delivery teams, data governance, cybersecurity, legal, privacy, and business leaders to embed governance requirements into AI design, deployment, and operations.
Enable adoption.
Promote consistent understanding of AI governance expectations through clear communication, practical guidance, and measurable stakeholder engagement across the governance community.

Benefits

Paid parental Leave
Adoption Assistance
Fertility Treatment
Childcare Discount
Nursing Mom Support
Employee Assistance Programs: free resources available 24/7 to you and your family in the areas of mental health, family life, and career and financial guidance
Wellness Programs: incentives for wellness activities, wellness spending account, programs for building healthy habits, virtual physical therapy for joint, back, and pelvic health, health management programs and more.
Diabetes Management Program
Pet insurance
Identity Theft Protection
PerkSpot Discount Program
Tuition assistance and career development programs