Explore Tasks
Classify an AI system's risk and assure its documentation, evaluation and oversight evidence
The AI system owner and the risk committee receive a recorded risk classification under the EU AI Act and internal policy, the controls and documentation the system needs, and periodic findings on whether its evaluation, monitoring and human-oversight evidence is current, all kept in the AI use register.
1job does it
5jobs help with it
52skills it needs
- AI Governance Lead
- AI Engineer
- Lead AI & ML Engineer
- Data Protection Officer
- Data Governance Manager
- Head of Data Governance
Does itHelps with it
- Reading and tracing unfamiliar code
- Reviewing human- and AI-generated code changes for correctness and risk
- Sampling and reviewing a team's work against standards
- Diagnosing misleading data visualisations
- Verifying reports and dashboards before and after release
- Verifying AI-assisted outputs against sources
- Disclosing AI contribution to professional work
- Testing competing explanations with discriminating evidence
- Independently checking analytical conclusions and claims
- Escalating unresolved risks with evidence and clear ownership
- Giving specific developmental feedback on data and AI work
- Judging the materiality of defects and deficiencies for the intended use
- Designing automated pipelines and processes that fail safely and recover
- Designing and running evaluations of AI systems
- Classifying AI uses against regulatory obligations
- Designing and appraising samples
- Enforcing and auditing data access
- Designing testable data controls
- Monitoring and testing data control effectiveness
- Mapping proposed process changes and control impacts
- Testing the completeness and accuracy of reports and data that controls rely on
- Explaining control types, dependencies and compensating controls
- Assessing data-service continuity and recoverability
- Judging whether and how to use AI for a task
- Evaluating fitness and full cost of AI-assisted work
- Analysing error patterns and failure modes in models and AI systems
- Monitoring model fitness and drift
- Evaluating responsible AI use risks
- Recognising personal data and privacy duties
- Assessing privacy risk through data protection impact assessments
- Recognising cybersecurity threats to data work
- Recognising and containing suspected personal data breaches
- Assessing data-driven fairness and bias
- Assessing supplier dependency, data-handling and exit risk
- Rating risks against appetite and choosing proportionate treatment
- Exercising human oversight of automated and AI outputs
- Specifying AI transparency and explanation duties to affected people
- Investigating anomalies
- Making accountable decisions under uncertainty
- Identifying and quantifying decision downside and exposure
- Revising prior decisions when evidence changes
- Reasoning about systems, feedback and second-order effects
- Facilitating group decisions using disputed evidence
- Recognising and resolving ethical conflicts in data and AI work
- Validating records by matching, recomputation and end-to-end tracing
- Tracing data lineage upstream and downstream
- Applying data classification and handling rules
- Assessing lawful basis, purpose compatibility and minimisation for data reuse
- Assigning and operating data ownership and stewardship
- Mapping personal data processing and flows
- Discovering undeclared data processing and AI use from operational evidence
- Record-keeping requirements for processing activities and AI systems
Paths that teach the skills it needs
Skill Path
Data Literacy for Business Decisions
Judge whether a business claim from a table, KPI, chart or dashboard really holds up.
- Modules
- 5 open
- Practice data
- Raket Parts
Skill Path
Data Preparation: Cleaning, Repair and Reconciliation
Check whether a table is fit for the job, then clean, repair and reconcile it with the evidence to back it.
- 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