Artificial Intelligence Trends and Strategy Training Course

Participants can explain major AI technology developments and their strategic implications.

• Machine learning, deep learning and foundation model capability boundaries
• Transformer architecture concepts: tokens, embeddings, attention and context windows
• Retrieval-augmented generation compared with fine-tuning and prompt engineering
• Multimodal models handling text, images, audio, video and documents
• Agentic systems using planning, tool calling, memory and human approval
• Compute, data and energy constraints shaping model deployment choices

Participants can produce a structured assessment of relevant AI trends and signals.

• Gartner Hype Cycle interpreted alongside evidence from operational deployments
• Horizon scanning sources: academic papers, standards bodies, releases and patents
• Signal-to-noise assessment separating capability evidence from marketing claims
• Scenario planning around model cost, regulation, competition and workforce effects
• Trend radar categories for technologies, vendors, skills and policy changes
• Assumption logs documenting uncertainty, confidence levels and review triggers

Participants can assess whether data, architecture and operating conditions support proposed AI use cases.

• Data readiness assessment covering quality, lineage, access, ownership and retention
• Information classification for personal, confidential, regulated and public data
• Knowledge base design for retrieval-augmented generation applications
• Data product principles defining producers, consumers, contracts and service levels
• Integration patterns using APIs, event streams, batch pipelines and workflow tools
• Readiness heat map scoring data, skills, controls, technology and sponsorship

Participants can identify AI use cases with clear users, decisions, constraints and intended outcomes.

• Process decomposition identifying repetitive tasks, decisions, handoffs and delays
• Customer journey analysis locating service friction and information gaps
• Decision inventory distinguishing assisted, automated and human-led decisions
• Use-case canvas defining users, inputs, outputs, controls and measures
• Failure mode analysis for incorrect outputs, omissions and unsafe actions
• Human-in-the-loop design specifying review thresholds and escalation routes

Participants can prioritise AI initiatives using transparent value, feasibility and risk criteria.

• Weighted scoring model combining value, feasibility, risk and strategic fit
• Impact-effort matrix applied to candidate AI initiatives and process changes
• Value stream mapping for cost, cycle time, quality and service benefits
• Dependency mapping across data, systems, vendors, policy and workforce
• Portfolio balance between quick wins, foundational work and strategic bets
• Stage-gate criteria for discovery, pilot, deployment and scale decisions

Participants can create an investment case for an AI initiative with measurable benefits and costs.

• Benefits hypothesis linking AI outputs to operational and financial measures
• Baseline measurement for productivity, accuracy, throughput, risk and satisfaction
• Total cost of ownership covering models, data, integration and support
• Sensitivity analysis for adoption rates, model usage and benefit assumptions
• Discounted cash flow concepts for multi-year AI investment proposals
• Benefits realisation plan assigning owners, measures, targets and review dates

Participants can identify, assess and treat key ethical, operational and model risks.

• NIST AI RMF functions: Govern, Map, Measure and Manage
• Risk taxonomy covering bias, hallucination, privacy, security and misuse
• Model cards documenting intended use, limitations, evaluation and monitoring
• Fairness testing methods for disparate impact and subgroup performance
• Red teaming prompts for harmful content, prompt injection and data leakage
• Residual risk decisions with accountable owners and documented acceptance

Participants can design governance structures, policies and assurance checkpoints for AI systems.

• ISO/IEC 42001 management system requirements for organisational AI governance
• AI policy clauses covering approved tools, data handling and prohibited uses
• RACI matrix assigning business, technical, risk and audit responsibilities
• Model inventory fields for ownership, purpose, data, risk and lifecycle
• Assurance gates for procurement, testing, release, monitoring and retirement
• Board reporting indicators for value, incidents, compliance and model performance

Participants can compare deployment architectures and sourcing options against organisational requirements.

• Build, buy and partner decisions using capability, cost and control criteria
• Public cloud, private cloud and on-premises deployment trade-offs
• Model gateway patterns controlling access, logging, routing and spend
• Vendor due diligence covering data rights, security, service levels and exit
• Open-weight and proprietary model selection against performance and obligations
• Reference architecture connecting identity, data, models, applications and monitoring

Participants can specify generative AI services that are useful, testable and appropriately controlled.

• User story writing for AI-assisted tasks and measurable acceptance criteria
• Prompt patterns for extraction, classification, drafting and structured reasoning
• Grounding strategies using retrieval, citations, rules and trusted sources
• Conversation design for clarification, confirmation, refusal and escalation
• Evaluation datasets representing typical, difficult and high-risk user requests
• Product metrics covering task success, trust, adoption and correction rates

Participants can determine where agentic AI and automation are appropriate and how to control them.

• Agent workflow design separating planning, execution, validation and approval
• Tool permissions restricted through least privilege and transaction limits
• Process orchestration combining robotic process automation, APIs and AI agents
• Exception handling for missing data, conflicting instructions and failed actions
• Audit trail requirements recording prompts, tools, outputs and approvals
• Autonomy levels matched to consequence, reversibility and human oversight

Participants can incorporate security and information protection requirements into AI strategy decisions.

• Threat modelling for prompt injection, model theft and data exfiltration
• OWASP Top 10 for LLM Applications applied to solution design
• Access controls using identity federation, role permissions and secret management
• Data loss prevention rules for prompts, attachments and generated outputs
• Secure development lifecycle controls for AI applications and integrations
• Incident response playbooks for harmful output, leakage and compromised models

Participants can define the operating model, roles and adoption actions required for sustained AI use.

• Target operating model covering decision rights, teams, processes and measures
• Federated AI delivery balancing central standards with business ownership
• Role design for product owners, model owners, data stewards and reviewers
• Skills matrix covering AI literacy, engineering, assurance and change leadership
• Change impact assessment for tasks, controls, workloads and workforce confidence
• Adoption plan combining communications, training, support and feedback channels

Participants can produce a sequenced AI roadmap with measurable milestones and review mechanisms.

• Roadmap sequencing across foundations, pilots, deployment, scaling and optimisation
• Milestone design linking deliverables, decisions, dependencies and accountabilities
• OKRs for AI value, adoption, reliability, risk and capability development
• KPI hierarchy connecting model metrics to process and business outcomes
• Portfolio review cadence for funding, benefits, risk and retirement decisions
• Executive dashboard design showing progress, exceptions, spend and realised value

Participants can communicate AI strategy choices clearly to executives, sponsors and oversight bodies.

• Executive narrative linking strategic objectives, choices, risks and investment
• Decision memos presenting options, evidence, assumptions and recommendations
• Visual portfolio maps showing priorities, sequencing, dependencies and ownership
• Board-level risk reporting with thresholds, incidents and treatment status
• Stakeholder mapping for sponsors, users, technical teams and control functions
• Challenge sessions testing strategic assumptions, trade-offs and decision criteria

Participants can build, review and present a complete AI strategy portfolio for a defined organisational context.

• Business context statement defining priorities, constraints, stakeholders and strategic outcomes
• Trend assessment identifying relevant technologies, uncertainties and planning assumptions
• Prioritised use-case portfolio supported by scoring, value and feasibility evidence
• Governance model specifying accountability, risk controls, assurance and policy requirements
• Implementation roadmap sequencing foundations, pilots, scale decisions and capability development
• Executive presentation defending investment choices, measures, risks and next actions