Solutions
About
News & Insights
Contacts
EN ▾
MAINS LAB

How to Oust a Top Manager?

In this article, you'll find tips to help you rise to the top in your company — and stay a decent human being — as well as reflections on the transformative power of AI in modern business management. These are ideas we believe are worth sharing.

Companies that fail to adapt today, that don’t explore new approaches to long-standing processes, risk becoming the next Kodak — which, despite having an employee who pitched the digital camera idea, chose to keep selling film.

So, what skills does a modern top manager need to avoid becoming a fossil?

Let’s look at the key competencies that define success in the age of AI — including career success.
If you're not a top manager yet, that’s fixable. To paraphrase modern economist Richard Baldwin:
“AI won’t replace you. But a manager who uses AI will.”
The world is full of industries — and no matter what people say, all industries are conservative. Once an industry forms, it quickly builds a shell: people set up a status quo and protect it with rules, standards, policies, and restrictions, and fiercely resist those who try to go around them.

But what if a new tool emerges? One that doesn’t break the rules, but simply makes their limitations irrelevant?

Take insurance, for instance — one of the most conservative sectors. Large players elbow each other for market share, trying to prove their service is more convenient than that of the competition. Then along comes a healthcare insurance startup that fully automates underwriting, drastically reduces claims processing time, enhances fraud detection, and improves customer satisfaction. Just four years after its founding, it hits one million customers. We wish we could say that company was Mains Lab — but it was another one: the U.S.- based startup Lemonade. Still Mains Lab is profitable, which is saying something.

What is Lemonade using? AI. They’ve embraced a data-driven approach — large-scale adoption of AI, computer vision, and machine learning for risk assessment.

Will Lemonade eventually go bankrupt due to flaws in its bold model? That’s not the point. What matters is that they’re trying to reshape the insurance market.
As Thomas Edison once said after another failed attempt at inventing the lightbulb:
"I have not failed. I've just found 2,000 ways that won't work."
Introduction
publication date:
April 7, 2025
TAGS:
SHARE:
Machine learning
Machine learning
Trends
Trends
Data is the new oil. But just like oil, it needs refining. Sure, extracting oil is difficult — but possible. Running an engine on crude oil? Impossible.

So the critical question is: Are you using crude or refined fuel? Before making data-driven decisions, make sure your data has been through the metaphorical refinery.

A manager who relies on intuition over data might be a visionary — but research shows that it often doesn’t cut it. As highlighted in HBR’s “The Dark Side of Visionary Leadership,” strategic changes only succeed when middle management believes in the data behind them.

So, in addition to vision, leaders must:
2. Understand the Value of Data
If futurist author Arthur C. Clarke had lived to see this day, he might say:
"Any sufficiently advanced AI is indistinguishable from magic."
But AI, first and foremost, is a tool. In the hands of a leader who understands its nature and applies it to real business challenges, it becomes capital — a value-generating asset.

Think of AI as a thousand-armed genie, brimming with modern knowledge, but like genies in old tales, it doesn’t think like a human. Its impact depends entirely on how well its owner understands its strengths and limitations — and on knowing what they want.

Is AI useful in insurance? Just look at the large number of staff in most insurance companies — it’s clear many roles could be automated. The industry leaders are already using AI to support underwriting decisions by analyzing numerous parameters and reacting to market shifts faster than any expert. They use it also to detect fraud more effectively than random spot-checks by medical or insurance auditors.

Replacing everyone with AI is a utopia — and even dangerous. As robotics pioneer Rodney Brooks noted:
“Most successful AI deployments have a human somewhere in the loop (perhaps the person they are helping) and their intelligence smooths the edges.”

Top managers don’t need to be developers. But they do need to understand AI on a high level:
  • What can AI tools do?
  • Where are they essential?
  • Are they cost-effective?
  • Where might they cause harm?
1. Know the Limits of AI's Usefulness

Assess the quality and representativeness of their data to avoid flawed assumptions

Track key AI metrics — because, like medication, side effects can be worse than the illness

Embrace data enrichment — integrating external data expands context and reduces blind spots

3

2

1

(infrastructure, training, expert hires)

(better product, higher revenue, reputation boost, market advantage)

What’s the potential positive impact?

5. Lead Cultural Transformation
Human nature is conservative. History confirms this — just ask the Luddites, who famously smashed weaving machines 200+ years ago.

When AI enters the workplace, employees may see it as a threat. And that’s normal. But true leaders don’t just implement tech — they reshape the team thinking. They build cultures of experimentation, learning, and digital literacy. They update KPIs to reflect digital transformation. And most importantly — they lead by example.

Understanding AI, knowing what fuels it, balancing gains and losses, taking responsibility, and driving progress — these are the skills a top manager needs to lead in today’s world.
4. Rethink the Familiar
AI doesn’t just bend business rules — it creates new markets and accelerates disruption in the old ones. Legacy companies must adapt. Being flexible isn’t just adopting new tools — it’s also knowing when to let go of outdated practices.

Top managers should routinely ask:
“What if our business looked nothing like it does today? What would it become?”
Poor AI implementation can result in financial losses, lawsuits, and PR disasters. Just look at UnitedHealth (2023), where the AI model "nH Predict" mass-denied insurance claims — and triggered legal action.

Top managers should regularly ask:
3. Balance Innovation and Risk

2

4

What’s the cost?

(market lag, competitive gap, financial decay)

What’s the cost of doing nothing?

What are the algorithmic risks?

(bias, unintended consequences)

3

1

Disclaimer #1: These tips are meant for patient, diligent, and attentive professionals who would likely earn their promotion anyway — this advice might just get them there faster

These aren’t all the possible tactics — but knowing them may help you earn a promotion or prove you’re right where you belong.
We all know stories of ambitious — yet questionable — colleagues rising to power. These cases aren’t always scandalous, yet they’re real.

If you lack something (luck, charisma, or whatever else), AI can become your advantage. Use it to compete smartly — or to defend your turf if you're already in the top ranks.

And if you’ve read this far, chances are you’re already asking the right questions — and prepared to remove the black boxes from your business.

Here are your six (unofficial) tactics:
The most essential skill of the 21st century is the same one humans needed 200,000 years ago:
the ability to learn and adapt.

The winners won’t be those who talk loudest about technology — but those who actually use it.

The question is:
Who will act first — you or your competitors?
Conclusion
If you skipped straight to this section, these tips may not help you — they require patience and attention. Still, it's all in your hands.
The Promised Career Tips

Disclaimer #2: If you start applying these tips, your goal might change along the way

Break strategies into hypotheses. Test them with AI. Sometimes, the company’s foundation isn’t stone — it’s just belief. Predict errors, propose fixes — become both oracle and savior

Spot weak links and look for “naked emperors”

3

Don’t share your productivity secrets

No comment.

6

5

AI is just tech. But knowing how it works is a pair of binoculars. You’ll spot opportunities others miss — simply because they don’t have the lens.

Study your edge

4

Unreliable, yes — but insanely knowledgeable. Shrink learning cycles from years to weeks. You won’t become a guru overnight — but you’ll get dangerously close.

Let AI be your mentor

Create the illusion of relentless effort

2

Use AI to scan data, flag issues, prepare slides, draft reports. Schedule emails for off-hours. Soon, people will think you never sleep.

Delegate routine tasks to AI, not to people

Especially when the results won’t be reviewed externally. Give your team the job of validating AI’s output — save time, build respect.

1