Concept Map

This map organizes the workshop into learning clusters. Each cluster contains concepts that should be read together.

1. Why AI is a leadership challenge

Leading Change Through AI: AI is used as a case for leadership under transformation, not just a technology topic.
Continuous Transformation: there is no stable off-ramp where leaders can wait for change to finish.
Exponential Technology and Linear Adaptation: technology accelerates faster than human and organizational adaptation.
Adoption Gap: capability becomes value only when adoption, trust, skills, workflows, and governance catch up.

LLMs as a Language Revolution: LLMs operate in the medium humans use to coordinate: language.
AI as a Prediction Machine: AI predicts likely continuations rather than understanding in the human sense.
Hallucination as Plausibility Optimization: plausible output is not the same as truth.
Jagged Frontier: AI can be brilliant at some difficult tasks and weak at some simple ones.
Transformer Architecture: the technical base that made modern LLMs possible.

Six Strategy AI Leadership Framework: the whole operating frame.
Delegate: remove load from repetitive, draining, low-risk work.
Co-Create: use AI to improve thinking, preparation, challenge, and simulation.
Innovate: create new value, personalization, learning loops, and services.
Tasks vs Jobs: AI changes task bundles, not jobs in a simple one-to-one way.
Do Not Create an AI Elevator Operator: redesign the system instead of automating outdated workflows.

Protect: preserve people, judgment, trust, capacity, dignity, and integrity.
Govern: define rules, roles, data boundaries, review points, and accountability.
Develop: build AI literacy, verification, judgment, workflow design, and metaskills.
NIST AI Risk Management Framework: a governance reference model.
Future Skills and Metaskills: the human capabilities that become more important.

From Turing Test to Agentic AI: the movement from intelligence tests to systems that act.
Autonomy Levels for AI Agents: a practical analogy for co-pilot, conditional autonomy, and specialized autonomy.
Hybrid Human-Agent Teams: leadership when people and agents share work.
Human Agency Scale: why technical feasibility does not automatically mean people want full automation.

Iceberg Concept: the visible tool layer is not the whole story.
Context as Differentiator: advantage comes from distinct context, expertise, tacit knowledge, and judgment.
People Process and Culture Value Equation: AI value depends heavily on the human and organizational layer.
LLM-Wiki: the compiled learning layer.
RAG: the query layer that can use the wiki and sources later.