Artificial Intelligence Strategy and Leadership Training Course

Participants can identify, assess and sequence AI use cases using transparent portfolio criteria.

• Use-case discovery interviews structured around process pain points
• Jobs to Be Done identifies customer and employee decision needs
• Value-effort matrix ranks candidate AI initiatives before detailed assessment
• Scoring criteria combine value, feasibility, risk and strategic fit
• Benefit hypotheses define baseline, target metric and accountable owner
• Portfolio heat maps expose dependencies, duplication and sequencing constraints

Participants can assess the technical foundations required for responsible AI deployment.

• Data readiness assessment covers quality, lineage, access and retention
• CRISP-DM phases clarify data preparation and model development responsibilities
• Training, validation and test datasets separated to prevent misleading performance
• Model selection weighs accuracy, explainability, latency, cost and maintainability
• Retrieval-augmented generation connects language models to approved organisational knowledge
• Human-in-the-loop design assigns review thresholds and escalation decisions

Participants can evaluate generative AI opportunities, architecture options and practical limitations.

• Large language model capabilities matched to defined business tasks
• Prompt engineering patterns improve reliability, structure and source grounding
• Hallucination controls combine retrieval, validation rules and human review
• Build, buy or configure decisions compare control, speed and cost
• Application architecture separates user interfaces, orchestration, models and data
• Token consumption and inference costs included in operating cost estimates

Participants can establish proportionate AI governance, risk controls and assurance arrangements.

• NIST AI RMF functions structure Govern, Map, Measure and Manage activities
• ISO/IEC 42001 management system concepts define AI accountability arrangements
• Risk taxonomy covers bias, privacy, security, safety and reputational harm
• Model cards document intended use, limitations, evaluation and monitoring requirements
• Impact assessments identify affected groups, foreseeable harms and control owners
• Escalation thresholds specify when models require review, suspension or retirement

Participants can design an operating model that assigns ownership across AI delivery and oversight.

• Target operating model defines processes, people, technology and governance components
• RACI matrix allocates sponsorship, product ownership, approval and assurance duties
• Federated and centralised AI teams compared against organisational demand patterns
• Product management practices connect user needs, releases and outcome metrics
• Skills matrix identifies capability gaps across business, data and risk roles
• Communities of practice establish reusable standards, patterns and peer review

Participants can evaluate AI investment cases and manage supplier-related delivery risks.

• Total cost of ownership includes data, integration, licences, monitoring and support
• Business cases separate direct savings, avoided losses and service-quality benefits
• Stage-gate funding links investment releases to evidence and risk reduction
• Supplier due diligence examines data handling, model provenance and security controls
• Contract clauses address ownership, audit rights, performance and exit arrangements
• Make-buy-partner analysis tests strategic control against delivery capability

Participants can plan controlled AI implementation from discovery through operational release.

• Discovery phase validates user needs, process constraints and data availability
• Minimum viable product scope defines decisions, users, controls and success measures
• Agile delivery cadences align experimentation with accountable governance checkpoints
• MLOps practices manage versioning, deployment, monitoring and model rollback
• Acceptance criteria include performance, fairness, security and usability thresholds
• Pilot design specifies sample population, duration, comparison and exit criteria

Participants can prepare users, measure outcomes and manage operational change after AI release.

• Stakeholder mapping identifies affected users, decision-makers and control functions
• ADKAR change model structures awareness, capability and reinforcement activities
• Process redesign removes duplicated handoffs after AI-assisted decision introduction
• Performance dashboards track adoption, accuracy, exceptions and business outcomes
• Monitoring plans detect data drift, model drift and emerging user harms
• Post-implementation reviews compare realised benefits against approved business cases

Participants can present a credible AI roadmap with priorities, governance and implementation milestones.

• AI strategy canvas connects ambition, priorities, capabilities, risks and measures
• Roadmap horizons sequence foundations, pilots, scaling decisions and operating integration
• Dependency mapping identifies data, architecture, policy and skills prerequisites
• Executive narrative frames choices, trade-offs, controls and expected outcomes
• Board reporting indicators cover investment, risk exposure, adoption and benefits
• Ninety-day action plan assigns owners, decisions, evidence and review dates