Level
0
No Adoption.
AI is available. Behavior doesn't change.
What it looks like
Tools are deployed and ignored. The status quo wins by default.
The Framework — PolyCognitive Leadership™
Five levels of AI adoption, and the leader required to move them.
A diagnostic for organizations absorbing AI faster than their leadership can adapt. Five levels describe where most are stuck. The sixth — the PolyCognitive Leader — is the profile that moves them. The term was first coined by Robert Blaga, and the framework is grounded in 200+ first-person interviews with CHROs and CEOs on enterprise AI adoption.
Level
1
Adoption Without Value.
People use AI. Outcomes don't move.
What it looks like
Activity goes up. Throughput stays flat. AI becomes productivity theater.
Level
2
Value With Risk.
AI creates lift. Risk slips in beside it.
What it looks like
Speed climbs while control quietly erodes. The wins outpace the governance.
Level
3
Safe Value, Bad Workflow.
Tasks improve. The system stays the same.
What it looks like
AI is bolted onto legacy flows. Patchwork wins. The operating model never gets rebuilt.
Level
4
Smart AI, Dumb Human.
The system works. The humans atrophy.
What it looks like
Judgment, skill, and curiosity erode quietly. The WALL-E paradox: a system that works only as long as nothing surprising happens.
Level
5
The PolyCognitive Leader.
Makes the system smarter without making the humans dumber.
The destination
Drives adoption. Captures value. Controls risk. Redesigns work. Preserves human judgment as AI scales. This is the destination — and the leadership profile most organizations now have to manufacture, fast.
The curriculum — what gets the org unstuck
Five capabilities, each on two tracks.
The diagnostic above names where organizations get stuck. The curriculum below names what a PolyCognitive Leader does to move them past each stuck-point. Every capability runs on two parallel tracks — a human-leadership move and an AI-leadership move. Run one without the other and the organization stalls.
- 01
Driving AI Adoption
Unsticks Level 0 — No Adoption
- Human
- Surface and resolve the real reasons people resist.
- AI
- Understand what's actually worth pushing the team toward.
Outcome
Leaders leave able to diagnose why their team isn't adopting and run the conversation that unblocks it.
- 02
Extracting Value from AI
Unsticks Level 1 — Adoption Without Value
- Human
- Hold people accountable for outcomes, not AI activity.
- AI
- Know where AI creates real value vs. where it's just theater.
Outcome
Leaders leave able to walk into any AI initiative and tell you if it's creating value or theater.
- 03
Managing AI Risk
Unsticks Level 2 — Value With Risk
- Human
- Make risk concrete and set rules people will actually follow.
- AI
- Know where the real risks live — data exposure, IP leakage, hallucination, quality.
Outcome
Leaders leave able to set AI boundaries their team will actually respect — without killing speed.
- 04
Redesigning Workflows
Unsticks Level 3 — Safe Value, Bad Workflow
- Human
- Help people release workflows their identity is built around.
- AI
- Architect new ways of working instead of bolting AI onto broken processes.
Outcome
Leaders leave able to redesign how their team works for AI — not just bolt tools on.
- 05
Preserving Human Edge
Unsticks Level 4 — Smart AI, Dumb Human
- Human
- Protect the skills AI can't replace by making people use them.
- AI
- Know where AI is fragile and humans must stay sharp.
Outcome
Leaders leave able to spot which human skills are atrophying on their team and intervene before it's too late.
What the research says
The locus of failure is human, not technical.
Three independent research streams converge on the same picture: enterprise AI doesn't fail at the model layer. It fails at the layer where people, processes, and leadership decide what to do with it.
- Finding 0110 / 20 / 70
Of the value AI creates: roughly 10% comes from the algorithms themselves, 20% from technology and data, and 70% from people, processes, and adoption.
BCG · Expanding AI's Impact with Organizational Learning
- Finding 02~70%
Share of companies reporting little or no measurable impact from their AI investments to date — a finding that has held steady year over year.
MIT Sloan Management Review · BCG, annual AI research
- Finding 03≈3×
Likelihood that AI leaders (vs laggards) have their CEO and senior team directly engaged in setting AI strategy. The top differentiator is non-technical.
McKinsey · The State of AI, annual report
Each finding points at a different layer — value capture, ROI, organizational engagement — and each lands on the same conclusion. The framework below is a working model for what the leadership layer specifically has to do.
“Most AI initiatives don't fail at deployment. They fail between Level 0 and Level 2 — and the leader is the bottleneck.”
Robert Blaga
Position in the literature
Where this sits among the emerging work on leading mixed human-AI organizations.
PolyCognitive Leadership sits alongside a growing body of 2025 work on managing AI agents and the leadership response to them — including Jarrahi and Ritala's principal-agent framework in *California Management Review* (July 2025), BCG's *Machines That Manage Themselves* (2025), the MIT Sloan Management Review × BCG analysis of *The Emerging Agentic Enterprise* (2025), and the NBER working paper by Weidmann, Xu, and Deming on measuring human leadership skills on AI teams (2025).
Those frameworks address the agent side — how to design, constrain, and govern individual AI agents and the systems they operate inside. PolyCognitive Leadership addresses the leader side — the capability profile a senior leader needs to make mixed human-AI organizations produce value rather than theater, and the diagnostic that places an organization on the path.
The two layers are complementary. Robert's framework is grounded in 200+ first-person interviews with CHROs and CEOs running enterprise AI adoption, conducted across two decades of leadership development practice. It is the practitioner counterpart to the principal-agent work — focused on the human leader's competencies, the organization's stuck-points, and the curriculum that moves it forward.
From framework to capability
The training program is how the leader profile gets built.
The framework names what every organization now has to manufacture. The 2-day PolyCognitive Leadership training is the practitioner curriculum that builds it — leaders working through real decisions, with AI in the room, against the levels.