AI can give you an answer in seconds. It can analyse more information than you could read in a lifetime. It can identify patterns you might never spot, challenge assumptions, model scenarios and help you explore a decision from multiple perspectives, and it is getting better remarkably quickly.
But there is one thing I don’t believe we should hand over to it.
Our judgement.
In fact, I believe the opposite is true. The more capable AI becomes, the more valuable human judgement becomes, because the challenge facing transformation leaders has rarely been simply a lack of information. The real challenge is deciding what to do when the answer isn’t obvious.
And that’s a very different problem.
Intelligence isn’t the same as judgement
Imagine you’re leading a major transformation.
The programme is six months behind schedule. The technology team believes another three months will solve the problem. The CFO is concerned about escalating costs. Operations wants to reduce the scope. Your CEO has already made commitments to the Board.
Meanwhile, your people are exhausted after several years of continuous change, and one of your most important executives privately tells you they no longer believe the transformation will deliver what was originally promised. Ask an AI system what you should do and it could probably give you an excellent analysis.
It could identify your options, assess risks, suggest scenarios, recommend questions and help you explore the consequences of different decisions. That’s enormously valuable, but eventually somebody still has to decide:
What are we actually going to do?
And that decision exists within a context that cannot always be reduced to data. There are personalities involved. History. Trust. Politics. Timing. Relationships. Organisational appetite. Personal credibility. Things that have been said in rooms that aren’t recorded anywhere. Things that haven’t been said at all, and sometimes are simply instincts developed from encountering similar situations many times before.
That’s where judgement enters the equation.
Transformation has never been purely rational
One of the central ideas behind the Transformation Leaders Body of Knowledge is that transformation needs to be understood as a system.
Strategy cannot be separated from people. Technology cannot be separated from operating models. Governance cannot be separated from decision-making. Culture cannot be separated from leadership behaviour, and change in one part of an organisation invariably creates consequences somewhere else.
That’s one reason transformation is so difficult.
Rarely is there a neat equation where we can enter the variables and find the right answer. We’re dealing with complex organisations made up of human beings. AI doesn’t change that. If anything, it makes recognising it even more important.
AI is removing one constraint, and exposing another
For much of my career, access to information has been a constraint. We commissioned research. Built business cases. Produced reports. Hired advisers. Ran workshops. We could spend weeks gathering and interpreting information before a leadership team was ready to make a decision.
AI is dramatically reducing that constraint. A leader can now interrogate extraordinary quantities of information almost instantly. That’s an incredible opportunity, but when information becomes abundant, the bottleneck moves.
The question is no longer simply: Can I get enough information to make this decision?
Increasingly, it becomes: Can I make sense of everything available to me, determine what actually matters and make a good decision?
That requires judgement, and I suspect it will become one of the defining leadership capabilities of the AI era.
The risk of outsourcing our thinking
We understandably spend a lot of time discussing AI hallucinations, bias, data quality and security. Those issues matter, but I think another risk gets less attention.
The gradual outsourcing of our thinking.
There is an important difference between using AI to help you think and using AI to think for you. The first strengthens leadership. The second could weaken it. If every difficult question immediately gets handed to an AI system, there is a danger that leaders gradually stop developing some of the capabilities they need most:
Critical thinking. Curiosity. Challenge. Pattern recognition. Reflection. Experience, and ultimately, judgement.
The irony would be extraordinary.
We could have access to more intelligence than any generation of leaders in history while becoming less confident in our own ability to make decisions.
This isn’t an argument for simply trusting your gut
There is an important distinction here. I’m not suggesting experienced leaders should ignore evidence and simply rely upon instinct. Experience can create wisdom.
It can also create bias. “I’ve seen this before” can be incredibly valuable.
It can also be completely wrong. Good judgment, therefore, isn’t about choosing between human experience and machine intelligence. It’s about using both intelligently.
AI can challenge our assumptions. Expose gaps in our reasoning. Find contradictory evidence. Present alternatives. Model scenarios. Ask questions we haven’t considered and force us to articulate why we believe something.
But the leader remains responsible for interpreting that intelligence within the context of their organisation.
Three forms of intelligence
This leads to a model that I believe will become increasingly important for transformation leaders.
We need to bring together three forms of intelligence.
Codified Intelligence — established knowledge, frameworks, research, methodologies and lessons accumulated over time.
Within TLH, much of this is captured within the Transformation Leaders Body of Knowledge.
Machine Intelligence — AI’s ability to analyse, synthesise, challenge, interrogate and explore information at enormous speed.
This is increasingly where Transforma-ai can support leaders as a 24/7 thinking partner.
And then there is Human Intelligence — experience, organisational context, relationships, empathy, peer insight and the tacit knowledge that often exists outside formal systems.
This is where leadership experience and the collective intelligence of communities such as TLH become particularly valuable. Bring those three together and something powerful happens.
Codified Intelligence + Machine Intelligence + Human Intelligence
create the environment for: Better Personal Judgement
which ultimately enables: Better Transformation Decisions.
None of the three forms of intelligence is sufficient on its own, and critically, none removes the leader from the equation.
AI should be the thinking partner, not the decision-maker
This is one of the principles behind the approach we’re developing within The Transformation Leaders Hub. I don’t think transformation leaders need another machine confidently telling them what the answer is. They need something that helps them think better.
Something capable of asking:
- Have you considered this?
- What evidence supports that assumption?
- What happens if you’re wrong?
- Whose perspective is missing?
- What are the second-order consequences?
- What would make you change your mind?
- What aren’t you discussing because it’s uncomfortable?
Those questions don’t remove human judgement. They strengthen it. That’s an important distinction in how we think about Transforma-ai. The objective isn’t to replace the transformation leader. It’s to make sure they don’t have to think through difficult transformation decisions alone.
Because leadership can be isolating
I’ve spent more than 25 years working around organisational change and transformation, and one thing I’ve increasingly recognised is how isolated transformation leaders can become. Not necessarily socially.
Decisionally.
The more senior you become, the fewer people there may be with whom you can openly explore uncertainty. Your team expects direction. Your executive colleagues have their own responsibilities and perspectives. External advisers may arrive with a particular solution or methodology. The Board wants confidence.
Yet you’re often dealing with situations where the truth is: “I’m not completely sure.”
That is where the combination of AI, codified knowledge and peer experience becomes particularly interesting. It creates somewhere to test assumptions, explore uncertainty and challenge thinking before the leader has to take a position.
That is very different from asking a chatbot for an answer. It’s about creating a better environment in which human judgement can operate.
From better answers to better decisions
This is the shift I think we need to make. We need to become less obsessed with getting answers from AI and more interested in using AI to improve decisions. Those aren’t the same thing.
A good answer can still lead to a poor decision if the context is misunderstood.
Equally, an imperfect analysis can contribute to an excellent decision when combined with experience, dialogue and judgement. So perhaps the question leaders should increasingly ask isn’t: “What does AI think I should do?”
It is: “How can AI help me think more clearly about what I should do?”
That small change creates a fundamentally different relationship with the technology. The leader remains responsible. AI becomes an input. Not the authority.
Transformation is a capability
This connects to another foundational belief behind TLH: Transformation is a capability, not simply a programme.
Organisations shouldn’t become good at transformation only while a major initiative is underway. They need to build an enduring capability to Diagnose, Decide, Execute, Reflect and Evolve. AI should become part of that capability, but so should judgement. So should leadership. So should organisational learning. So should the ability to challenge assumptions and change direction when the evidence tells us something isn’t working.
The objective shouldn’t be an organisation where AI makes more decisions. It should be an organisation capable of making better decisions because AI is available. Those are very different ambitions.
The differentiator won’t be who has the best prompts
There will undoubtedly be a period where knowing how to operate particular AI tools creates an advantage, but tools change. Models improve. Interfaces become easier. Capabilities that appear extraordinary today will eventually become commonplace. So I don’t believe knowing how to operate AI will ultimately be the differentiator.
Knowing when to trust it, when to challenge it, when to ignore it and when something important is missing might be.
That’s judgment. And judgment isn’t downloaded. It’s developed through experience, reflection, challenge, mistakes, conversations and exposure to different perspectives.
Which brings us back to something decidedly human.
AI may give transformation leaders access to extraordinary intelligence, but intelligence alone doesn’t lead transformation.
People do.
Ultimately, somebody still has to look at the evidence, understand the context, accept the uncertainty and say:
This is what I believe we should do next — and here’s why.
AI doesn’t make that responsibility disappear. It makes our ability to exercise it well even more important.
