AI is only half the equation.
Technology is the easy half. The half that turns AI into lasting value — and builds a company that stays successful — is human: intent, trust, alliance, leadership. Measured in value, not in sentiment. That’s the half I work on.
It isn’t a technology problem.
Eight independent studies from 2025–2026 point the same way: where AI stalls, the barrier is rarely the model. It’s the human half no one budgeted for.
show no measurable P&L impact — they stall before they scale.
yet workflow redesign is the #1 driver of AI’s bottom-line impact.
of companies never reshaped jobs around AI — new tools, old work.
of leaders: built-in human intervention will always be crucial to AI.
tech delivers ~20% of an initiative’s value; the other 80% is redesigning work.
Eight independent studies, one finding: the missing half is human.
Sources: MIT Project NANDA; BCG; Stanford HAI AI Index; McKinsey State of AI; KPMG Global AI Pulse; Deloitte State of AI; EY AI Pulse; PwC AI Predictions. The 95% figure is an early, self-reported signal — cited as direction, not settled fact.
−19pp
A controlled Harvard experiment — not a survey.
When 758 BCG consultants used AI inside its capability frontier, work quality rose 40%. Outside it, they were 19 points more likely to be wrong. The gain isn’t automatic — it depends entirely on how humans wield it.
“AI won’t replace humans — but humans with AI will replace humans without AI.”
It looks like a tech problem. It almost never is.
Failure starts before the model.
The most common single root cause of AI project failure is human: teams miscommunicate what problem the AI should solve, so it’s built for the wrong one.
Tools no one trusts go unused.
Adoption runs ahead of trust, and trust is context-dependent — not innate hostility. The work is earning it, not overriding it.
New tools, old workflows.
84% of companies bolt AI onto unchanged work. Value shows up only when the work itself is redesigned around the human–machine team.
Four dimensions, one core.
The lens I use with clients and across the series: four human dimensions on two axes — mindset vs. motion, self vs. together — that decide whether AI creates value. Close them, and the value follows.
Why, before the tool
Naming the real problem AI should solve — before anyone buys anything.
The real adoption curve
Earning the trust of people — and leaders trusting them — that decides whether AI gets used.
What the machine can’t lead
Carrying people through the shift — and the agile freedom to act — no algorithm leads for you.
One team, not two halves
Tech-enabled work where humans and AI collaborate as one — the whole, not the halves.
Value · Intent · Trust · Alliance · Leadership
VITAL — the precondition for lasting success: what it takes for humans and AI to achieve something together, and make it hold.
Value isn’t a slogan here: 77% vs 20% business outcomes by talent confidence (KPMG); workflow redesign the #1 driver of EBIT (McKinsey); +40% work quality when humans wield AI well (Harvard).
Three moves to make it whole.
Where is your half missing?
An honest read of where people, leadership and culture — not the stack — are blocking adoption.
Build the human half.
The trust, capability and leadership an AI shift actually runs on — not slideware, real readiness.
Make it whole.
Bring both halves together — the method beneath the brand. The whole is more than the sum.
Three decades closing that gap.
I spent 30+ years at Arthur Andersen and EY, most of them as a Senior Partner, leading enterprise and SAP-driven transformations across industries. The pattern never changed: the technology was solvable; the human half decided the outcome.
“As part of the EY leadership behind the firm’s transformation research with Oxford’s Saïd Business School, I saw the same pattern hold up close: organizations that put humans at the center were far more likely to succeed.”
The Missing Half — the series.
The series is live. The argument for the human half of AI, one piece at a time — each along one of the four dimensions above, each anchored in the evidence, every figure with its source. Read the series · Why I’m doing this.
I’m not alone at this conclusion.
The world’s leading management thinkers — the 2025 Thinkers50 — converge on one message: organizations win when people flourish.
Psychological safety is the foundation of teams that dare to adopt, experiment and fail forward with AI.
“Co-intelligence”: keep the human in the loop — humans and AI as one collaborating team, not tool versus user.
The “Turing Trap”: building AI to replace people is a choice — augmenting them creates far more value.
Catching an inflection like AI early is decided by human agility, not by the technology itself.
Generative AI is the first technology to truly hit knowledge work — so work must be redesigned around people.
Innovation is deeply social — leaders unlock collective genius; the AI era needs more of it, not less.
“Artificial integrity”: AI has to be built to serve human values — the human-centered case in one phrase.
Trust is the foundation of leadership — and of getting people to adopt anything new, AI included.
Future-ready organizations are the ones whose people can adapt — human adaptability sets the pace, not the tech.
Category winners and Top-10 of the Thinkers50 2025 ranking. Cited as independent convergence — not endorsement.
Even the people building AI agree.
The leaders of the labs racing to build AI keep landing in the same place: the machine is the easy part — the human who wields it decides the value.
You’re not going to lose your job to an AI — but you’re going to lose your job to someone who uses AI.
“We want to build tools to augment and elevate people, not entities to replace them.”
Frames AI as a powerful collaborator — not a replacement for human ingenuity.
The honest exception: some foresee AI and robots replacing most work entirely, making it optional. A horizon worth taking seriously — but today the evidence is clear: value is captured where humans and AI work together, not where humans step out.
Two halves. One whole.
AI brings the half that computes. People bring the half that matters. Put them together — that’s the work.
Start the conversation.
Advisory, board work, or a keynote that makes a room rethink its AI strategy — tell me what you’re facing.