Phase 1: Honestly place yourself
The roadmap starts with the self-assessment from "The Four AI Competency Levels — and Where You Stand Today": are you at AI-Aware, AI-Enabled, or already AI-Fluent? This honest placement decides which of the following phases deserve priority.
Phase 2: Train judgment as the foundation
From "Reading AI Outputs Critically: the Actual Core Skill" follows the second phase: turn the three check questions (source traceability, visibility of possible errors, stakes) into a fixed habit before delegating more tasks to AI at all. This phase is the foundation for everything that follows.
Phase 3: Delegate deliberately and request precisely
The third phase connects two modules: "Co-Pilot, Not Autopilot: Delegating the Right Tasks" defines what's delegable and what stays human responsibility, "Prompting as a Project Leadership Skill" supplies the craft for it – give context, set format, provide an example, iterate.
Phase 4: Secure the data foundation and communication
The fourth phase guards against the two most common failure sources: "Data and Bias Fundamentals: Why the Input Data Decides Everything" ensures AI-assisted forecasts rest on a solid foundation, "Translating AI Limits Honestly: the Communication Skill" ensures that foundation is presented to stakeholders neither overstated nor understated.
Practice section: Phase 5 – staying current
The last phase isn't a one-time close-out, but a recurring routine from "Keeping Up With the Pace of AI Development": review your own tool choices at a regular interval and deliberately update outdated knowledge. This exact routine decides whether the competency built up in the first four phases stays at the AI-Fluent level – or is already outdated again a year from now.