AI Governance Lead
Accountable for assessing the risk of AI uses and for the AI management system and model documentation that show how AI is governed.
4tasks in this job
20skills it uses
6tools named
The skills this job leans on most
- 1Classifying AI uses against regulatory obligationsNeeded in 4 of 4 tasks
- 2Evaluating responsible AI use risksNeeded in 3 of 4 tasks
- 3Discovering undeclared data processing and AI use from operational evidenceNeeded in 2 of 4 tasks
- 4Record-keeping requirements for processing activities and AI systemsNeeded in 2 of 4 tasks
- 5Specifying AI transparency and explanation duties to affected peopleNeeded in 1 of 4 tasks
ToolsOneTrustCredo AIMLflowLangfuseJiraMicrosoft ExcelMicrosoft PurviewBigIDServiceNowMicrosoft SharePoint
All 4 tasks, where people get stuck and all 20 skills
Tasks
- Evaluate an AI use case for suitability, value and risk before adoption
- Classify an AI system's risk and assure its documentation, evaluation and oversight evidence
- Set and review the organisation's AI use policy and AI literacy requirements
- Maintain the AI use register and AI management system evidence across all AI systems
Where people get stuck
- Designing enforceable data governance policies
- Maintaining a risk register with owners, triggers and responses
- Diagnosing capability gaps from work evidence and planning development
All skills
- Evaluating responsible AI use risks
- Record-keeping requirements for processing activities and AI systems
- Applying data classification and handling rules
- Assessing supplier dependency, data-handling and exit risk
- Assigning and operating data ownership and stewardship
- Designing and appraising samples
- Designing testable data controls
- Escalating unresolved risks with evidence and clear ownership
- Explaining control types, dependencies and compensating controls
- Judging the materiality of defects and deficiencies for the intended use
- Judging whether and how to use AI for a task
- Mapping proposed process changes and control impacts
- Monitoring and testing data control effectiveness
- Rating risks against appetite and choosing proportionate treatment
- Recognising and containing suspected personal data breaches
- Recognising cybersecurity threats to data work
- Recognising personal data and privacy duties
- Testing the completeness and accuracy of reports and data that controls rely on
- Tracing data lineage upstream and downstream
- Validating records by matching, recomputation and end-to-end tracing
Paths to start
Tool Path
Excel for Analysis and Decision Models
Build and check Excel workbooks others can trust.
- Modules
- 5 open
- Runs in
- Your browser
Skill Path
Responsible AI Governance and Assurance
Govern and assure AI across its lifecycle including third-party AI and redress.
- Modules
- 5 open
- Practice data
- Raket Parts
Skill Path
Privacy, Security and Access for Data and AI
Protect personal data and access with impact assessments and incident evidence.
- Modules
- 5 open
- Practice data
- Raket Parts
7tasks in this job
60skills it uses
10tools named
The skills this job leans on most
- 1Classifying AI uses against regulatory obligationsNeeded in 6 of 7 tasks
- 2Evaluating responsible AI use risksNeeded in 5 of 7 tasks
- 3Specifying AI transparency and explanation duties to affected peopleNeeded in 2 of 7 tasks
- 4Designing enforceable data governance policiesNeeded in 1 of 7 tasks
- 5Maintaining a risk register with owners, triggers and responsesCore to this role
ToolsOneTrustCredo AIMLflowLangfuseJiraMicrosoft Excel+ 4 more
All 7 tasks, where people get stuck and all 60 skills
Tasks
- Evaluate an AI use case for suitability, value and risk before adoption
- Prepare a model or AI system release decision with evaluation evidence, monitoring and rollback
- Maintain a data and AI risk register and record risk treatment and acceptance
- Classify an AI system's risk and assure its documentation, evaluation and oversight evidence
- Oversee an automated or AI system in operational use and correct, restrict or withdraw faulty outputs
- Set and review the organisation's AI use policy and AI literacy requirements
- Maintain the AI use register and AI management system evidence across all AI systems
Where people get stuck
All skills
- Classifying AI uses against regulatory obligations
- Designing enforceable data governance policies
- Evaluating responsible AI use risks
- Analysing error patterns and failure modes in models and AI systems
- Applying data classification and handling rules
- Assessing data-driven fairness and bias
- Assessing data-service continuity and recoverability
- Assessing lawful basis, purpose compatibility and minimisation for data reuse
- Assessing privacy risk through data protection impact assessments
- Assessing supplier dependency, data-handling and exit risk
- Assigning and operating data ownership and stewardship
- Bounding and securing AI systems' tool and data access
- Building data protection by design into data products and pipelines
- Controlling disclosure risk in shared data and outputs
- Decomposing one's own tasks for AI assistance
- Designing AI-assisted workflows with bounded tools and human authority
- Designing and appraising samples
- Designing and running evaluations of AI systems
- Designing automated pipelines and processes that fail safely and recover
- Designing decision rights and dispute routes for data and AI work
- Designing testable data controls
- Discovering undeclared data processing and AI use from operational evidence
- Enforcing and auditing data access
- Escalating unresolved risks with evidence and clear ownership
- Evaluating ML generalisation and fitness
- Evaluating fitness and full cost of AI-assisted work
- Exercising human oversight of automated and AI outputs
- Explaining control types, dependencies and compensating controls
- Gating releases on evidence with rollback ready
- Governing shared data definitions
- Identifying and prioritising critical data elements
- Interpreting policy scope and delegated authority limits
- Investigating anomalies
- Judging the materiality of defects and deficiencies for the intended use
- Judging whether and how to use AI for a task
- Locating and acting on personal data for individual rights requests
- Mapping personal data processing and flows
- Mapping proposed process changes and control impacts
- Mapping stakeholders' decisions, constraints and needs
- Monitoring and testing data control effectiveness
- Monitoring model fitness and drift
- Presenting and defending evidence-based recommendations to decision-makers
- Preserving incident evidence and timeline
- Protecting sensitive data in AI-assisted work
- Rating risks against appetite and choosing proportionate treatment
- Recognising and containing suspected personal data breaches
- Recognising cybersecurity threats to data work
- Recognising personal data and privacy duties
- Recognising statutory incident reporting duties and deadlines
- Record-keeping requirements for processing activities and AI systems
- Resolving consent and suppression status across systems
- Running incident command: severity, timeline and status updates
- Setting and applying a release and rollback bar for models and AI systems
- Setting and reviewing action thresholds for scores and model predictions
- Testing the completeness and accuracy of reports and data that controls rely on
- Tracing data lineage upstream and downstream
- Training reproducible machine learning models
- Validating records by matching, recomputation and end-to-end tracing
- Working through stakeholder conflict over evidence and responsibility
- Writing data quality rules
Paths to start
Tool Path
Excel for Analysis and Decision Models
Build and check Excel workbooks others can trust.
- Modules
- 5 open
- Runs in
- Your browser
Skill Path
Responsible AI Governance and Assurance
Govern and assure AI across its lifecycle including third-party AI and redress.
- Modules
- 5 open
- Practice data
- Raket Parts
Skill Path
Privacy, Security and Access for Data and AI
Protect personal data and access with impact assessments and incident evidence.
- Modules
- 5 open
- Practice data
- Raket Parts
8tasks in this job
83skills it uses
12tools named
The skills this job leans on most
- 1Classifying AI uses against regulatory obligationsNeeded in 6 of 8 tasks
- 2Judging whether and how to use AI for a taskNeeded in 5 of 8 tasks
- 3Evaluating fitness and full cost of AI-assisted workNeeded in 5 of 8 tasks
- 4Evaluating responsible AI use risksNeeded in 5 of 8 tasks
- 5Specifying AI transparency and explanation duties to affected peopleNeeded in 2 of 8 tasks
ToolsOneTrustMLflowMicrosoft PurviewJiraMicrosoft PowerPointMicrosoft Power BI+ 6 more
All 8 tasks, where people get stuck and all 83 skills
Tasks
- Evaluate an AI use case for suitability, value and risk before adoption
- Prepare a model or AI system release decision with evaluation evidence, monitoring and rollback
- Prepare a periodic performance or risk report for a management or governance committee
- Maintain a data and AI risk register and record risk treatment and acceptance
- Classify an AI system's risk and assure its documentation, evaluation and oversight evidence
- Oversee an automated or AI system in operational use and correct, restrict or withdraw faulty outputs
- Set and review the organisation's AI use policy and AI literacy requirements
- Maintain the AI use register and AI management system evidence across all AI systems
Where people get stuck
- Classifying AI uses against regulatory obligations
- Judging whether and how to use AI for a task
- Evaluating fitness and full cost of AI-assisted work
All skills
- Classifying AI uses against regulatory obligations
- Evaluating fitness and full cost of AI-assisted work
- Evaluating responsible AI use risks
- Judging whether and how to use AI for a task
- Analysing error patterns and failure modes in models and AI systems
- Applying data classification and handling rules
- Assessing data-driven fairness and bias
- Assessing data-service continuity and recoverability
- Assessing lawful basis, purpose compatibility and minimisation for data reuse
- Assessing privacy risk through data protection impact assessments
- Assessing supplier dependency, data-handling and exit risk
- Assigning and operating data ownership and stewardship
- Bounding and securing AI systems' tool and data access
- Building data protection by design into data products and pipelines
- Building decision-focused reports and dashboards
- Communicating quantified uncertainty
- Communicating sensitive or unwelcome analytical findings
- Constructing clear and honest charts
- Controlling disclosure risk in shared data and outputs
- Decomposing one's own tasks for AI assistance
- Designing AI-assisted workflows with bounded tools and human authority
- Designing and appraising samples
- Designing and running evaluations of AI systems
- Designing automated pipelines and processes that fail safely and recover
- Designing decision rights and dispute routes for data and AI work
- Designing enforceable data governance policies
- Designing testable data controls
- Diagnosing misleading data visualisations
- Disclosing AI contribution to professional work
- Discovering undeclared data processing and AI use from operational evidence
- Enforcing and auditing data access
- Escalating unresolved risks with evidence and clear ownership
- Estimating quantities and differences with intervals
- Evaluating ML generalisation and fitness
- Evaluating and improving chart accessibility
- Exercising human oversight of automated and AI outputs
- Explaining control types, dependencies and compensating controls
- Explaining technical work to accountable non-specialists
- Gating releases on evidence with rollback ready
- Giving specific developmental feedback on data and AI work
- Governing shared data definitions
- Identifying and prioritising critical data elements
- Independently checking analytical conclusions and claims
- Interpreting correlation without inferring cause
- Interpreting policy scope and delegated authority limits
- Investigating anomalies
- Judging the materiality of defects and deficiencies for the intended use
- Judging whether a change exceeds normal variation
- Judging whether uncertainty changes the decision
- Locating and acting on personal data for individual rights requests
- Mapping personal data processing and flows
- Mapping proposed process changes and control impacts
- Mapping stakeholders' decisions, constraints and needs
- Monitoring and testing data control effectiveness
- Monitoring model fitness and drift
- Presenting and defending evidence-based recommendations to decision-makers
- Preserving incident evidence and timeline
- Protecting sensitive data in AI-assisted work
- Rating risks against appetite and choosing proportionate treatment
- Reading charts and dashboards correctly, including their state
- Recognising and containing suspected personal data breaches
- Recognising and resolving ethical conflicts in data and AI work
- Recognising cybersecurity threats to data work
- Recognising misleading numerical claims
- Recognising personal data and privacy duties
- Recognising statutory incident reporting duties and deadlines
- Record-keeping requirements for processing activities and AI systems
- Resolving consent and suppression status across systems
- Running incident command: severity, timeline and status updates
- Sampling and reviewing a team's work against standards
- Selecting charts for evidence and decisions
- Setting and applying a release and rollback bar for models and AI systems
- Setting and reviewing action thresholds for scores and model predictions
- Testing competing explanations with discriminating evidence
- Testing the completeness and accuracy of reports and data that controls rely on
- Tracing data lineage upstream and downstream
- Training reproducible machine learning models
- Validating records by matching, recomputation and end-to-end tracing
- Verifying AI-assisted outputs against sources
- Verifying reports and dashboards before and after release
- Working through stakeholder conflict over evidence and responsibility
- Writing data quality rules
- Writing evidence-led findings and auditable recommendations with their limits
All tools
- OneTrust
- MLflow
- Microsoft Purview
- Jira
- Microsoft PowerPoint
- Microsoft Power BI
- Microsoft Excel
- Credo AI
- Microsoft SharePoint
- ServiceNow
- Langfuse
- Varonis
Paths to start
Tool Path
Excel for Analysis and Decision Models
Build and check Excel workbooks others can trust.
- Modules
- 5 open
- Runs in
- Your browser
Skill Path
Responsible AI Governance and Assurance
Govern and assure AI across its lifecycle including third-party AI and redress.
- Modules
- 5 open
- Practice data
- Raket Parts
Skill Path
Privacy, Security and Access for Data and AI
Protect personal data and access with impact assessments and incident evidence.
- Modules
- 5 open
- Practice data
- Raket Parts
Coming soonApplied AI at Work
About the role
Accountable for assessing the risk of AI uses and for the AI management system and model documentation that show how AI is governed. Decides each use's risk classification under the EU AI Act and internal policy and judges whether its training data, evaluation and human oversight are adequate. Leads this without line management, where a Data Governance Lead owns data quality and access and the Lead AI & ML Engineer sets technical release standards.
Also called AI Risk Lead, Responsible AI Lead, AI Governance Manager (when the role has no team)