Why AI Transformation Is a Leadership Problem, Not a Technology Problem
Created on 2026-02-06 09:22
Published on 2026-02-26 09:45
The uncomfortable truth that executives avoid and organizations pay for
I delivered 241% sales growth at Electrolux Malaysia.
Then I was fired.
My Chairman, Gunnar Broberg, looked at me and said words that changed the trajectory of my career: “Indhran, you are brilliant. But you are not ready for management.”
At the time, I did not understand. The numbers were extraordinary. KPMG had audited them. The results were real.
But Gunnar saw something I could not see. I was optimizing for metrics while destroying the organization. I was brilliant at output but toxic in the hallway. I was getting results that could not be sustained because I had neglected everything that makes results sustainable.
I was leading without actually leading.
This failure taught me something that applies directly to AI transformation today. The technology is not the constraint. Leadership is the constraint.
And most executives do not want to hear this.
The Comfortable Lie
There is a comfortable lie that executives tell themselves about AI failure.
The technology is not ready. The vendors oversold capabilities. The integration was too complex. The data was not clean. The timing was wrong.
These explanations are comfortable because they locate the problem outside the executive. The failure happened to us. We were victims of technology, vendors, or circumstance.
The uncomfortable truth is different.
AI transformation fails because leaders fail to lead it.
Not fail to approve it. Not fail to fund it. Fail to lead it.
MIT’s research found that 95% of organizations get zero return from AI investments. When they investigated why, they did not find technology failures. They found:
“This divide does not seem to be driven by model quality or regulation, but seems to be determined by approach.”
The approach is set by leadership. The failures are leadership failures.
The Delegation Trap
The most common leadership failure in AI transformation is what I call the Delegation Trap.
The pattern is predictable.
The CEO recognizes AI is important. The board is asking questions. Competitors are making announcements. Something must be done.
The CEO approves budget. Hires a Head of AI or assigns responsibility to the CIO. Announces the initiative. Returns attention to “real” priorities.
“Keep me updated.”
This feels like leadership. It is not.
What happens next:
The Head of AI builds a team. Runs pilots. Generates activity.
But the Head of AI cannot align the executive team. Only the CEO can do that.
The Head of AI cannot remove organizational obstacles. Data owners who refuse to share. Functions who feel threatened. Leaders who quietly resist.
The Head of AI cannot resolve strategic tensions. Is AI primarily about cost reduction or capability building? Different executives have different views. Nobody has forced alignment.
The initiative accumulates friction. Progress slows. Enthusiasm fades.
Eventually, the CEO asks why more has not been achieved. The answer is always some version of “organizational barriers.” The CEO nods, expresses disappointment, and moves on.
The initiative dies. The Delegation Trap claims another victim.
Why executives fall into the trap:
AI feels technical. Leaders without technical backgrounds assume technical people should handle technical matters.
This is a category error.
AI transformation is not a technical matter. It is an organizational matter. The technology works. The organization does not change.
Transforming organizations is exactly what executives are supposed to do. It is their core job. Delegating it to technical staff is abdicating responsibility.
The Understanding Requirement
Let me be direct about what leadership requires.
Leaders do not need to understand how AI technology works. They do not need to explain neural networks, transformer architectures, or training methodologies.
They need to understand what AI changes about their business.
This is a different kind of understanding. It is strategic, not technical. It answers different questions.
What AI changes:
How will competitive dynamics shift? What advantages will AI create or eliminate? What happens to organizations that lead versus organizations that lag?
What decisions will change? Which human decisions will AI support, augment, or replace? What new decisions become possible?
What capabilities are required? What must the organization learn to do? What skills become more valuable? What skills become less valuable?
What risks emerge? Where can AI fail in ways that damage the business? What governance is needed? What human judgment must be preserved?
The articulation test:
Can you articulate, in two sentences, what AI will change about your competitive position?
Not generic statements about efficiency and innovation. Specific claims about specific changes.
If you cannot pass this test, you are not ready to lead AI transformation. You are ready to approve it and delegate it. Which means you are ready to fail.
The Personal Adoption Signal
Here is something many executives do not want to hear.
If you are not personally using AI, you are not credibly leading AI transformation.
Organizations watch what leaders do, not what they say. When leaders announce AI initiatives but never use AI themselves, the organization receives a clear message: this is important for others, not for leaders.
The Shadow AI Economy that MIT identified, where 90% of workers use personal AI tools while only 40% of companies have official subscriptions, tells a story.
Workers are adopting AI. Many leaders are not.
This creates a credibility gap. Leaders are asking organizations to change while not changing themselves. The organization notices.
What personal adoption demonstrates:
When leaders use AI visibly, they demonstrate that AI is safe. If the CEO uses it, it cannot be career-limiting.
When leaders discuss their AI experiences, they normalize conversation. Others share their own experiences. Learning accelerates.
When leaders make mistakes with AI and acknowledge them, they create psychological safety. Others feel permitted to experiment and fail.
The minimum bar:
Leaders do not need to become power users. The minimum bar is regular, visible use of AI for actual work.
Using AI to draft communications. To analyze information. To explore options. To prepare for meetings.
This is not difficult. AI tools have become remarkably accessible. The barrier is not technical skill. The barrier is willingness.
Leaders unwilling to clear this bar are signaling that AI is not actually a priority. The organization responds accordingly.
The Alignment Imperative
AI transformation requires executive team alignment. Not polite agreement in meetings. Actual alignment on direction.
Where misalignment hides:
Strategic tensions often remain unstated. Different executives have different visions for AI, but nobody forces the conversation that would surface the conflict.
The CFO sees AI as cost reduction. Automate tasks. Reduce headcount. Improve margins.
The CRO sees AI as revenue growth. Better customer insights. Faster sales cycles. New offerings.
The CHRO sees AI as capability building. Workforce development. Talent retention. Future readiness.
Each view is legitimate. But they imply different priorities, different investments, different success metrics.
When these tensions remain unresolved, the organization receives conflicting signals. Initiatives optimized for cost reduction conflict with initiatives optimized for revenue growth. Resources are contested. Progress stalls.
The CEO’s job:
Only the CEO can resolve these tensions.
The Head of AI cannot tell the CFO that cost reduction is secondary. The CIO cannot tell the CRO that revenue growth takes priority. These are executive decisions that require executive authority.
The CEO must force the conversation. Surface the tensions. Make choices. Communicate direction clearly.
This is uncomfortable. Executives do not enjoy being told their priorities are secondary. But unresolved tensions are more costly than uncomfortable conversations.
The Commitment Ceremony:
I use a technique I call the Commitment Ceremony.
At the end of alignment discussions, I ask for explicit individual commitments.
“I am asking each of you to commit to supporting this initiative for the next 90 days. This includes providing resources when requested, removing obstacles when they arise, and not pulling the plug based on early results that may look poor due to normal transformation dynamics. Can you make that commitment?”
Go around the room. Get individual commitments. Document them.
This feels awkward. Awkwardness is the point. Explicit commitment is more durable than implicit assumption.
Leaders who cannot make the commitment reveal misalignment that needs to be addressed. Leaders who make the commitment have created accountability they will feel.
The Obstacle Removal Function
Leaders exist to remove obstacles that others cannot remove.
AI transformation encounters obstacles constantly. Data owners who will not share. Functions who feel threatened. Legacy systems that resist integration. Policies that prohibit necessary actions.
Most of these obstacles cannot be removed by the AI team. They require authority that the AI team does not have.
The escalation pattern:
The AI team encounters an obstacle. They try to resolve it at their level. They fail because the obstacle involves people or decisions beyond their authority.
They escalate. If leadership is engaged, the obstacle is removed. If leadership has delegated and disengaged, the escalation goes nowhere.
The obstacle remains. The initiative slows. Eventually, accumulated obstacles stop progress entirely.
What engaged leadership looks like:
Engaged leaders are accessible when escalation is needed. They understand the initiative well enough to evaluate obstacles quickly. They have relationships that enable resolution.
When the data team will not share data, the engaged CEO calls the Chief Data Officer. The conversation happens. The data flows.
When a function resists change, the engaged CEO addresses the resistance directly. Either the resistance is resolved or the resistor is moved.
This is not micromanagement. It is obstacle removal. It is the leadership function applied to the specific context of AI transformation.
What disengaged leadership looks like:
Disengaged leaders are unavailable or slow to respond. They do not understand the initiative well enough to evaluate obstacles. They defer decisions that require their authority.
The AI team escalates. Days pass. Weeks pass. Nothing happens.
The obstacle remains. The team works around it, compromising the initiative. Or the team gives up, and the initiative dies.
The disengaged leader may never know what killed the initiative. They only know it failed.
The Electrolux Lesson Applied
Let me return to where I started.
At Electrolux, I was brilliant at results and terrible at leadership.
I optimized for the spreadsheet. I ignored the relationships, the culture, the sustainability of what I was building.
The results were real. The organization was breaking.
AI transformation can fail the same way.
Leaders can optimize for AI metrics. Pilots launched. Models deployed. Adoption statistics reported.
While the Human Layer breaks.
Trust erodes as people fear displacement. Culture degrades as experimentation is punished. Capability gaps widen as training is superficial. Processes break as AI is layered on dysfunction.
The metrics look good. The organization is failing.
The sustainable path:
AI transformation that lasts is not faster or more aggressive. It is more complete.
It addresses leadership alignment before demanding organizational change. It builds capability before expecting performance. It redesigns processes before automating them. It creates governance before deploying at scale.
This is slower than metric-optimized approaches. It is also more likely to succeed.
The 5% who succeed with AI are not moving faster than the 95% who fail. They are moving more deliberately. They are building foundations that enable results to sustain.
The 18-Month Decision
The 18-month window I have written about is a leadership decision.
Organizations that build AI readiness now create compound advantages. Data advantages. Capability advantages. Competitive positioning that compounds over time.
Organizations that wait face growing barriers. Competitors’ head starts. Talent scarcity. The window closing.
This is a leadership decision. Not a technology decision. Not an operational decision. A decision about strategic priorities and resource allocation.
What the decision requires:
Understanding the stakes. Not from vendor presentations, but from genuine grasp of how AI is reshaping competitive dynamics.
Willingness to prioritize. AI transformation requires resources. Those resources come from somewhere. Something is deprioritized when AI is prioritized.
Commitment to lead personally. Not approve and delegate. Lead. Be present. Remove obstacles. Maintain alignment.
Tolerance for the J-curve. Early results will likely disappoint. The initiative must be protected through the valley.
The question:
The question is not whether your organization will adopt AI. The question is whether leadership will lead the adoption or delegate it.
Delegation has a 95% failure rate. Leadership offers something better.
What Leadership Actually Looks Like
Let me be concrete about what AI transformation leadership looks like in practice.
Weekly engagement:
Leaders allocate time weekly to AI transformation. Not monthly updates. Weekly engagement.
This might be a standing meeting with the AI team. It might be reviews of progress and obstacles. It might be personal use of AI tools being deployed.
The frequency signals priority. Weekly says this matters. Monthly says it is routine.
Obstacle removal velocity:
When obstacles are escalated, they are resolved within days, not weeks.
This requires leaders to understand the initiative well enough to act quickly. Leaders who need extensive briefing before deciding are too slow.
Fast obstacle removal maintains momentum. Slow obstacle removal kills it.
Visible championing:
Leaders talk about AI transformation publicly and frequently.
In all-hands meetings. In communications. In conversations with teams.
They share their personal experiences with AI. They celebrate progress. They acknowledge challenges.
Visibility signals priority. Silence signals indifference.
Strategic integration:
AI appears in strategic discussions, not just operational reviews.
How does AI affect competitive positioning? How should AI inform strategic planning? How does AI change the assumptions underlying strategy?
When AI is only operational, it remains peripheral. When AI is strategic, it becomes central.
AI transformation is not a technology problem. It is a leadership problem.
The technology works. Vendors deliver capability. Implementation is possible.
What fails is leadership. Delegation instead of engagement. Misalignment instead of clarity. Obstacles that remain instead of obstacles removed.
This is uncomfortable for executives to hear. The failure is not external. The failure is theirs.
But recognizing the problem is the first step to solving it.
AI transformation succeeds when leaders lead. Not approve. Not fund. Not delegate.
Lead.
How engaged is your leadership with AI transformation? What would change if engagement increased?
The AI Readiness Scorecard assesses your leadership dimension alongside five other dimensions of the Human Layer. It takes ten minutes and shows where readiness gaps exist.
Comment “SCORECARD” below and I will send you access.
The technology is ready. The question is whether your leadership is ready.