What happens when a decision you made fails? Revenue falls, stocks crash, and the worst part is that employees who trusted you have to be laid off.
A leader’s decision-making has to be strategically sound. Like a game of Chess or Go, even if it can’t predict the future, the move must make sense in your own context. Whether that’s serving the shareholders, improving employee engagement, or developing something that truly matters.
Yes, revenue matters. Without it, businesses are doomed.
However, it’s a decision that leaders make that leads to revenue, and there are a lot of decisions to go around. A leader is swamped and cognitively spent every day.
That’s why C-suites, investors, and managers protect themselves so closely- because a wrong decision, at their level, means loss of their job and their subordinates. Whether someone admits it or not, that is a blow to a person’s pride as a leader.
However, there is a golden line: AI is helping leaders make decisions. But the reason behind it might not be that the machine knows better.
The reason is simple, it’s a library that can be questioned and if properly used can question back. And yet, it can become a crutch to be relied on. If the AI is taking all decisions, what are you doing?
Being accountable won’t be enough. The tool must be harnessed the way the steam engine was- with deliberate precision.
The Chessboard: Strategic Choices in the Age of AI
In Chess, once the players make 5 moves each, the number of possible moves is an unimaginable number called the Shannon Number, named after Claude Shannon. The number is 69,352,859,712,417
Businesses function similarly; your decisions branch out into many possibilities. It’s basic probability. And it is possibly one of the first things any leader learns: no decision is 100% foolproof. Something always goes awry.
Doesn’t that sound exhausting?
The Weight of Too Many Moves: Decision Fatigue.
Decision fatigue is a real phenomenon experienced by leaders of all kinds. The scope hardly matters; the number of decisions does, and the scale.
After all, you can’t delegate these. And anyone doing so is no leader but rather a caricature of the typical office manager. While actions can be delegated, decisions cannot. And with decisions comes the noise of the people surrounding you.
“Where are the reports?”
“Hey, can you help the XYZ team with the product planning?”
“What about Q2? What have you done to increase revenue for the upcoming quarter?”
“Yeah, Tom from product got fired. We need a new hire, and you need to sign off on it. Oh, what do you mean you didn’t know?”
And this is in the first 4 hours of you being in the office. A leader, it seems, has to put out fires. Not to mention the onslaught of salespeople calling you to sell their solution to you. Luckily, you got a PA to handle that.
Wait, where’s your PA?
Another task, another day.
All leading to analysis paralysis.
From Grandmaster to Game Piece: Risk Management or Complacency?
AI can change all of that. This powerful tool offers everything a leader needs, from simulation to acting as a PA to screening calls to automating workflows.
Wow. The potential is endless.
You can finally break the chain of analysis paralysis and make informed decisions. Isn’t that why you got those AI systems in the first place? To augment and not reduce?
But what happens when you become complacent and dependent? Finally, the decisions have been outsourced, but their weight hasn’t.
You still carry them. The consequences are still with you.
This article by Forbes illustrates how top leaders, more than their employees, are using ChatGPT for better answers- for better actionable insights. The article says that 84% have bought products based on ChatGPT recommendations.
This is inevitable.
But what if the products are bad and your leadership is questioned?
What if it’s good, you over-rely on AI and subconsciously begin delegating decision-making, only passively approving, rather than being an active strategist?
The question you have to ask here is: who is making the move?
The Machine as Second Board: Using AI Without Losing Agency.
That’s enough pessimism. But let’s look at this: AI isn’t just threatening the jobs of your subordinates. But leaders as well.
It is a reality you will eventually have to face.
But what if you didn’t have to?
One of the most under-talked events in corporate history is how Shell overcame the 1973 crisis.
They just fed their top execs with possibilities. Shell hired analysts who used statistics to help leaders think in novel ways. And what do you know? Shell avoided one of the most perilous oil crashes in history.
This is what AI is for.
It won’t play for you, but it will help you understand the probabilities of what your move might do. And then empower you to pick the best one. Leaders must experiment with this technology- using it broadens what they think is possible.
But the use cases with which they are currently being used are almost unimaginative. It’s not used to see the possibilities but to eliminate them. Hasn’t the unpredictability of the stock market shown you that prediction is impossible and everything is an educated guess?
The future of leadership does not rest with the businessman who can predict. But instead, one who can create possibilities, something AI systems are primed for.
The Battlefield: Leadership, Risk, and Accountability
When Moves Fail: Accountability
AI systems can make your business feel safe. But the cascading effects of this tech are still unknown. Just speculation that AI might be a bubble sent analysts and the market scrambling. If the tech fails to live up to its almost magical promise, the market will burst.
And leaders who championed this in-house will be blamed.
Millions lost to a “safe bet.”
This is accountability. Not to mention the loss of jobs that any leader will have to carry with them.
However, AI is genuinely utilitarian. Between touting it as a medicine for all woes and an enemy of the people, there is a middle path that everyone is overlooking: AI is technology, not an evolution.
It was the natural course that our computing powers were going to take, as Alan Turing intended. And the failure or success of this tech will be attributed to the leader using it.
Let’s illustrate this.
Say you buy an Agentic AI, you bought it because one of your AI bots suggested it after days or months of talking and comparing. All the things that Google used to do.
But the confidence you have in this decision won’t be the same as Google’s because there, the opinions were of human beings, prone to bias. You have collaborated with the LLM to actually be authentic.
And the agentic AI fails.
Who do you blame? Yourself or the machine?
That’s a question to think about.
When Moves Succeed: Leadership
But what if it works and proves ROI? This is the scenario we want.
And not just once, mind you, but many times. The decisions work as you intended. And perhaps this is where the tech is going, making decisions foolproof.
But then why do shareholders need the CEO, the CFO, the CMO, or anyone? It could just be a web of AIs talking to each other and making necessary decisions.
Isn’t this a practical question to consider? Virtually, small businesses would be eliminated by this because what a small business could do, a $5000 AI system could do, too. And the future of that monopoly seems likely.
A lot of AI-agents that were specialised tools will be cannibalized by players with more resources.
What then is the role of the leader?
It is to move from decision-making to exploration. If your revenue is generated by machines, you will have to redirect the talented people in your employ to work on the possibilities the AI systems have put forth before you.
That is the logical, actionable step here.
The Library: AI as Advisor, Not Oracle
The Library That Questions Back
One of the best things about the AI is its ability to question when prompted. For example, if you have used Claude, it has a function to change its voice from learning to explanatory, to concise.
The explanatory and learning functions are amazing. They ask thought-provoking questions that help a leader reflect on what they’re doing.
It asks you questions like a teacher or someone who has expertise in that area. Of course, it’s not hallucination-free, but it doesn’t have to be to make someone think.
You guide it, and it guides you back.
That is a core difference between AI-dependent and AI-augmented decision-making.
The active participation of the leader.
One of the ways AI can help you and not cripple you is to find the blind spots that exist in your systems, like an expert would- but the AI system would be customized to your organization down to its micro-functions and then suggest changes.
But it will be the leader who will have to understand and implement the changes and ask the system why as many times as possible.
The possible hiccup
This is a sugarcoat. It’s not a hiccup but the elephant in the room- the visceral reaction many in top management had with AI: let’s replace our team.
Makes business sense to remove your programmers, marketing team, customer success, and whatnot, keeping the key members and integrating AI that might be cheaper to run in the long run.
Of course, this way, no one has to deal with the humanity of it all.
And this is a risky move. Not because people aren’t irreplaceable, but rather that failure won’t be able to be attributed, nor would anyone be held accountable- the loop would crush the organization.
The future has no merit because while thinking (AI!), action (robotics!), agency (Agentic!) can be substituted, observing real-time changes is a gift that only people possess. They can look at patterns and say- hmm, maybe this won’t work.
Great leaders don’t predict; they explore what is possible and what isn’t, and direct their attention there.
Bad leaders want to eliminate this because it’s too much for them.
The question is: Which one are you?
