4. Design the training

Teach people to do their real work differently, not to learn about AI in general.

U.S. Department of Labor AI Literacy Framework — summaries from Parker Poe and SHRM

Published in February 2026, it sets out five content areas and seven delivery principles for AI literacy training. It treats AI literacy as a baseline skill for every worker, not a technical specialty. A solid foundation for the "everyone" layer of any program.

The 6 Essential Skills Your Workforce Needs to Keep Pace with AI — Harvard Business Review / Gartner

Argues against treating AI skills as an ever-growing checklist. Instead, it focuses on a small set of capabilities in two areas: orchestrating work with AI, and overseeing it. Useful for designing training that holds up as the tools change.

PI approach: build the habit, not the course

• Train by role, using people's own tasks.

• Make verification the core skill: check every number, date and commitment.

• Give managers their own track. They make or break adoption.

• Protect the time. In a utilization-driven business, learning that competes with billable hours loses.