LEVEL 0045%
Pre-Adopter
From pre-chat, looking for use cases to refining my prompts & generally getting familiar.
AI Adoption Level Guide
This adoption canvas sets the success path for change-management — how people progress with AI and compound their systems.
The Game (Competency)
Leverage × efficiency × effectiveness — how much real value each person produces with AI.
The Path (This Rubric)
What people ship, how they climb, how the team compounds.
Adoption Note: Someone at L1 can produce high-leverage work on a specific problem. Someone at L4 can build agents that don't move the company forward. We want to climb the rubric together — empower people to create a bigger impact than they could on their own.
How to use the Level Guide: Pick your honest baseline, aim one level up, pick one small change you can repeat this week — not a leap. Top levels are very rare, earlier levels will give you a stronger base. We are all learning to progress together.
Maturity vs. Progress
REALISED PROJECTED EMPLOYEES
Personal
Team
Agents
ORGANISATIONAL PROGRESS
Click a status badge to move NOW. Click a card to see details. Right-click (or long-press) NEXT ▾ to set arrival estimate.
LEVEL 0149%
One-Off AI • Capable
People use AI one-off — copy-paste between tools. Gains real but individual; nothing builds on itself.
LEVEL 025%
Shared Knowledge • Accelerator
AI plugs into the company brain. Context flows from real systems, not copy-paste; the team queries shared knowledge.
LEVEL 031%
Shared Skills • Architect
The team builds with AI — purpose-built skills the team reaches for unprompted, multiplying output without headcount.
LEVEL 04
Autonomous Agents • Catalyst
Agents act on their own — running scheduled and event-driven workflows that advance goals without prompts.
LEVEL 05
Cross-Team Compounding • Multiplier
Agents run critical work flows & build new systems driving outcomes across teams. Impact per employee compounds at scale.
LEVEL 01 · NOW
One-Off AI • Capable
Types a question, reads the answer, goes back to working the way they already worked.
WHO THEY ARE
- Chat: Get advice.
- Smarter search: opens ChatGPT in a browser tab when stuck, & copy-pastes between tools.
- Often uses AI to research topics more efficiently — consistently saves time.
- Output looks about the same as pre-AI, but faster to first draft. A few chat tabs; maybe a system prompt.
GRADUATION CRITERIA
- Personal OS: You have a functional system that compounds knowledge & personal learnings for measurably better outputs, AND AI is connected to at least one real work system so it can reach your work without you pasting it in.
CLIMBING ACTIONS
- 10 min · Write a 1-paragraph “who I am / how I want responses” and save it where the AI loads it every session (CLAUDE.md, a Claude Project, a Copilot agent instruction, a Gemini gem). Test the same question with vs. without it.
- Bring AI to the work instead of bringing your work to the AI. Move into an AI-native surface as your start point: CoWork, Claude/Codex, an IDE, or Copilot inside the M365 app you already work in.
- Connect AI to at least one real work system (Slack, Drive, the CRM, Snowflake) — an MCP or connector, Copilot wired to SharePoint/Outlook, or Gemini in Workspace.