In a business environment where artificial intelligence is often discussed more than it is deployed, Gaurav Oberoi, Chief Digital Auditor, AI Trainer and Business Coach, occupies a pragmatic middle ground. An engineer by training and marketer by profession, Oberoi has spent more than 15 years scaling startups, e-commerce brands and legacy enterprises across India and the Middle East. His core expertise is around Digital Marketing, Technology, ecommerce and AI. Today, his focus is singular: helping organisations convert AI from a fashionable narrative into operational advantage.

A former regional trainer at Google across MENA and India, and a collaborator for TikTok. He has trained professionals from various regional and global organizations. Oberoi launched AI School to train executives to deploy effectively use AI to optimize the processes, deploy AI agents, automate core business functions and design measurable growth systems.

Having trained more than 9,000 professionals and advised over 300 companies, his thesis is direct: AI should reduce complexity, not increase it and growth should be engineered, not improvised.

In this conversation with Danish Shaikh, Editor at The International Wire, Oberoi discusses AI-enabled growth systems, leadership readiness, and why the real differentiator in the next decade will not be access to tools, but the discipline to use them well.

---

Beyond the Hype: Is AI Actually Replacing People?

You often say AI isn’t taking jobs—people who know how to use AI are. Is that a skill gap problem or a structural economic shift?

It’s both, and it’s easy to see why. Right now, we have a skill gap. Think about two graphic designers. One uses an AI tool to generate twenty different logo ideas in ten minutes. The other spends all day creating two by hand. The first designer is going to be far more valuable. That’s the skill gap in action.

But if you zoom out, you see the bigger picture. The economy itself is changing. For a long time, we valued people who could perform repetitive tasks reliably. Now, software can do that. The new jobs, the valuable jobs, are the ones that require creativity, strategic thinking, and a human touch. AI is just the tool that’s forcing this change to happen much faster than we expected. The skill gap is just a symptom of this much larger shift.

When a manager is replaced by automation, is that efficiency—or organisational failure?

It’s an organizational failure, without a doubt. If a manager’s entire job can be done by a piece of software, then they weren’t really a manager to begin with. They were an administrator. Their job was to check boxes, approve timesheets, and pass reports up the chain. That’s not leadership.

Real leadership is about coaching your team through a tough project. It’s about noticing a junior employee has potential and mentoring them. It’s about making a tough call when the data is ambiguous. You can’t automate wisdom or empathy. So, if a company replaces a “manager” with an algorithm, it’s a sign they never understood what real leadership was in the first place. They get a small efficiency gain, but they reveal a much bigger problem in their culture.

Are companies deploying AI strategically, or simply chasing cost compression?

Most companies are initially drawn in by the promise of saving money. It’s the easiest thing to understand and measure. “If we automate this process, we can reduce our headcount by 10%.” It’s a simple, compelling story for a CFO. And for some businesses, that’s as far as it goes.

But the smart ones, the ones who will win in the long run, see past the cost savings. They’re asking a more interesting question: “How can we use this technology to create something entirely new?” They’re not just trying to do the same things cheaper; they’re trying to do new things they couldn’t do before. For example, a bank might use AI not just to automate fraud detection, but to create truly personalized financial advice for its customers. Chasing cost savings is a defensive move. Using AI to create new value is how you win the game.

Do you believe the fear narrative around AI is overstated—or still underestimated?

The Hollywood-style fear narrative about sentient robots taking over the world is wildly overstated. It’s a fun topic for a movie, but it’s a distraction from the real issues. The dangers we should be talking about are far more practical and immediate.

I believe the real risks are still underestimated. For example, the speed at which AI is disrupting the job market is faster than anything we’ve ever seen. Our education systems and social safety nets are not prepared for that. Another huge risk is algorithmic bias. If we train AI on data that reflects our past biases, we’re not just continuing those biases; we’re automating them and scaling them up. The conversation needs to shift from science fiction to the real-world challenges of making sure AI is used fairly and responsibly.

Is AI genuinely creating new value pools, or redistributing existing ones?

It’s definitely creating new value. Of course, some industries are being disrupted, and value is shifting around. That’s always happened with new technology. Think about how the internet disrupted traditional newspapers. That was a redistribution of value.

But AI is also making it possible to solve problems we could never solve before, and that creates entirely new markets. Look at drug discovery. Scientists are using AI to identify potential new medicines in a fraction of the time it used to take. That’s not just redistributing money from old drugs to new ones; it’s creating a whole new potential for health and well-being. Or consider personalized education. AI can act as a personal tutor for every child on the planet. The economic and social value of that is immense. AI is making the entire economic pie bigger, not just slicing it up differently.

Enterprise Reality: Why AI Pilots Stall

Many companies experiment with AI but struggle to scale it. In your experience, what actually breaks?

What breaks is the jump from the lab to the real world. In the lab, everything is perfect. You have a clean, beautiful dataset, and your AI model works like a charm. But then you try to use it in the actual business, and it all falls apart.

Suddenly, you realize the data you need is spread across ten different old systems that don’t talk to each other. The employees who are supposed to use the new tool hate it because it’s clunky and they weren’t trained properly. The real world is messy, and the AI model wasn’t built for the mess. Scaling isn’t about making the AI smarter; it’s about cleaning up your own house – your data, your processes, and your culture – so the AI can actually do its job.

Is AI adoption primarily a technology challenge—or a leadership discipline problem?

It’s a leadership problem, 100%. The technology is the easy part. We have amazing AI tools available to us right now. The hard part is having the leadership to use them properly.

What does that look like? It means the CEO has to have a clear vision for how AI will help the company win, not just a vague idea that “we should be using AI.” It means having the guts to move your best people and your money into these new projects, even if it means starving some old, comfortable ones. And it means having the patience to stick with it when the first few experiments fail. Too many leaders are treating AI as an IT project they can delegate. It’s not. It’s a fundamental change to the business, and the leaders have to lead it from the front.

Where do you see the biggest gap between AI ambition and execution?

The gap is in middle management. I call them the “frozen middle.” The executives at the top are excited about the big vision, and the tech teams on the ground are excited to build cool new things. But the middle managers are stuck. They’re the ones who have to actually make it happen, but they’re often the most resistant.

Why? Because they’re judged on this quarter’s numbers, and this new AI project feels like a distraction. They’re worried it will make their job irrelevant. They don’t really understand the technology, and they’re being asked to lead a change they don’t believe in. So, the big vision from the top never makes it to the front lines. To fix this, you have to get your middle managers on board. You have to train them, empower them, and make them the heroes of the story, not the victims.

How should boards measure ROI from AI initiatives?

Boards need to think about it like a personal investment portfolio. You wouldn’t put all your money into one type of stock, and you shouldn’t measure all your AI projects the same way.

You should have three “buckets.” The first is for the safe bets, the “efficiency plays.” This is using AI to automate simple processes to save money. The ROI here is easy to calculate. The second bucket is for the “growth plays.” This is using AI to create new products or services that will make you more money. The ROI is measured in new sales or market share. The third bucket is for the long shots, the “strategic options.” These are the experimental projects that might not pay off for years, but if they do, they could change the entire game for you. The return here isn’t a number; it’s a new capability, a new market, a future. A smart board knows it needs a mix of all three.

What does a realistic 12-month AI transformation roadmap look like?

A realistic roadmap isn’t about changing everything at once. It’s about starting a flywheel. You start small, get a few quick wins, and build momentum.

• First 3 months: You figure out where to start. You find a few problems that are big enough to matter but small enough to solve. You get your data in order. You’re basically just getting organized.
• Next 6 months: You run your first few pilot projects. The goal is to get a real win, something you can show off. You want to be able to go to the rest of the company and say, “Look, we used AI to solve this problem, and it saved us a million dollars.”
• Last 3 months: You take your successful pilot and you scale it up. You also start telling everyone about your success. You’re not just scaling the technology; you’re scaling the excitement. At the end of the year, you haven’t transformed the whole company, but you’ve started a movement.

AI as Competitive Advantage

You speak about “AI-enabled growth systems.” What differentiates an AI-enabled company from one merely using tools?

It’s the difference between having a smart ingredient and having a smart recipe. A company that just uses AI tools is like a cook who buys a fancy truffle but doesn’t know what to do with it. They might grate it on top of a random dish. It’s okay, but it’s not transformative.

An AI-enabled company has a recipe. They build a system where everything is connected. For example, an e-commerce site uses AI to recommend products. It watches what you click on and learns what you like. That makes the recommendations better for you. But it also tells the company what products are popular, so they know what to stock more of. That data might even give them ideas for new products to create. It’s a smart cycle where the business is constantly learning and getting better. That’s a growth system. It’s much more powerful than just using an AI tool to write emails.

Does AI improve margins—or does it commoditise advantage?

It does both, just at different times. At first, using AI is like having a secret weapon. It helps you work faster and cheaper, which is great for your profit margins. But soon, all your competitors get the same secret weapon. It’s not special anymore. It just becomes the new standard, the cost of entry.

The real, long-term advantage comes from something that your competitors can’t copy. It’s not the AI tool itself, but how you use it. Do you have a unique brand that people love? Do you have a massive, loyal customer base that gives you data no one else has? Do you have a special way of doing things that AI makes even better? The AI becomes a booster for your unique strengths, and that’s an advantage that can last.

How do smaller firms compete when large enterprises have more capital to invest in AI?

They compete by being faster and smarter. A big company is like a giant cruise ship. It’s powerful, but it takes five miles to turn it around. A small company is like a speedboat. It can change direction in an instant. It can try a new idea on Monday, see if it works by Wednesday, and change it by Friday. A big company can spend a year just debating the idea.

Plus, the game has changed. You don’t need to spend millions building your own AI anymore. You can rent world-class AI from companies like Google, Amazon, or OpenAI for a fraction of the cost. So, the speedboat can have the same powerful engine as the cruise ship. In this world, it’s not about the size of your budget; it’s about the quality of your ideas and how fast you can execute them.

Is speed now more important than scale?

Absolutely. Speed is the new scale. For a long time, the goal was to be the biggest. The biggest factory, the biggest office, the most employees. But now, being big can actually be a disadvantage because it makes you slow. The companies that are winning today are the ones that can learn and adapt the fastest.

Think about a company like Instagram. It wasn’t the biggest photo company in the world when it started. But it was the fastest to understand what people wanted – to share photos easily from their phones – and to build a product that did it beautifully. The ability to learn what your customers want and give it to them before anyone else is the most powerful advantage you can have today. That’s speed. That’s the new scale.

What sectors in MENA are structurally best positioned for AI-led productivity gains?

The MENA region has a huge advantage because in many areas, it’s building from scratch. It doesn’t have to worry about old, outdated systems. It can jump straight to the future. Three areas come to mind:

• Finance: Instead of dealing with old-fashioned banks, we can build a financial system for the mobile-first generation. Imagine getting a business loan approved on your phone in minutes because an AI can instantly analyze your business plan and financials. That’s the opportunity.
• Retail: The e-commerce boom is just getting started. AI can make it so much better. Think of a grocery delivery service that knows you’re about to run out of milk and has it at your door before you even have to ask. That’s the kind of hyper-personalization AI can deliver.
• Energy: The region is a global leader in energy, and AI can make it a leader in energy efficiency too. Imagine using AI to predict exactly when a solar panel needs to be cleaned to get the most power from the sun, or to prevent a pipeline leak before it ever happens. It’s about making our most important industry smarter and more sustainable.
• Healthcare: is underserved in many parts of the region. AI-driven diagnostics, telemedicine, and predictive health interventions can scale access dramatically. I coached a healthtech company to use AI to read radiology scans in remote areas where specialists are scarce. That’s not incremental improvement—that’s transformational access
• Government Services: are also ripe for transformation. Reducing bureaucracy, improving citizen experience, and automating routine administrative tasks. Smart governments here are investing heavily in AI to deliver services faster and more transparently. The UAE’s government efficiency programs are genuinely among the most advanced I’ve seen globally.

The Workforce Question

If AI replaces 60–80% of repetitive roles, what happens to middle management?

Their job is going to change completely. The old job of a middle manager was to be a human spreadsheet – to track progress, write reports, and tell people what to do. AI can do all of that now, and do it better. So that part of the job is going away.

But the important part of the job, the human part, is becoming more critical than ever. The new job of a middle manager is to be a coach. It’s to help your team members solve tough problems, to support them when they’re struggling, and to create a culture where people feel safe to be creative. You’re no longer a manager of tasks; you’re a leader of people. It’s a harder job, but it’s also a much more meaningful one.

Should companies retrain or restructure?

You have to do both, but you have to do them in the right order. You retrain first, then you restructure. If you start by firing people, you create a culture of fear, and you lose all the valuable knowledge those people have about your business. It’s a short-sighted and destructive approach.

The smart way is to invest in your people. Give them the training they need to work with the new AI tools. As they learn new skills and start working in new ways, you’ll naturally see that the old company structure doesn’t make sense anymore. You’ll move from rigid departments to more flexible, project-based teams. The new structure will grow out of your newly skilled workforce. It’s a much more positive and effective way to change.

What skills will survive this cycle of automation?

The skills that will be most valuable are the ones that make us human. It’s the stuff that AI is terrible at. You can break it down into three main areas:

1. Solving Fuzzy Problems: AI is great at solving problems with clear rules, like a game of chess. But it’s terrible at solving problems where you don’t even know what the real problem is. The ability to think critically, to be creative, and to solve messy, real-world problems will be a superpower.
2. Working with People: As AI handles more of the technical work, the human work becomes more important. This is about empathy, communication, and collaboration. Can you inspire a team? Can you negotiate a difficult deal? Can you understand what a customer really wants, even when they can’t articulate it? These are deeply human skills.
3. Learning and Adapting: The most important skill of all might be the ability to learn. The world is changing so fast that your current skills will have a shorter and shorter shelf life. The people who succeed will be the ones who are constantly curious, who are always learning new things, and who can adapt to whatever comes next.

Are we underestimating the psychological cost of AI-driven restructuring?

Completely. We talk a lot about the people who lose their jobs, but we don’t talk enough about the people who are left behind. They are living in a constant state of anxiety. They’re worried their job is next. They’re stressed about having to learn new, complicated systems. It’s exhausting, and it kills creativity and morale.

Leaders have a huge responsibility to manage this with care. You have to be honest with people. You have to give them a clear vision of where the company is going and how they fit into that future. You have to give them real training and support, not just a link to a website. You have to create a culture where it’s okay to be a little scared and to ask for help. The stress people are feeling is real, and good leaders take it seriously.

How should governments respond—education reform, regulation, or market adaptation?

Governments need to do all three. There’s no single magic answer.

• Long-Term: Fix Education. Our schools are still teaching kids to be good at things that computers are already good at, like memorizing facts. We need to start teaching them how to be creative, how to think critically, and how to work together. This is the most important thing we can do for the next generation.
• Medium-Term: Help People Adapt. In the meantime, we need to help the people who are already in the workforce. This means more support for job training programs, and a stronger social safety net (like unemployment benefits and healthcare) to help people land on their feet when their job is eliminated.
• Short-Term: Smart Rules. We need some rules for AI, but they have to be smart. They should be focused on things like preventing bias and protecting our privacy. The goal isn’t to slow down technology, but to make sure it’s moving in a direction that benefits everyone, not just a few big companies.

AI in the Gulf Context

How does AI adoption in the UAE and MENA differ from India or Europe?

It’s a completely different mindset. In the UAE and the Gulf, the government is in the driver’s seat. There’s a national plan, a big vision that says, “We are going to be a world leader in AI, and this is how we’re going to do it.” It’s strategic and it’s happening at a huge scale.

In Europe, it’s more cautious and fragmented. Individual companies are doing interesting things, but they’re often worried about regulations like GDPR. In India, you have this incredible, vibrant ecosystem of tech talent, but it’s often focused on providing AI services to other countries. The Gulf’s approach is unique because the government is trying to build the entire future economy from the top down, with AI at its core.

Is the Gulf moving faster because it has less legacy infrastructure to protect?

That’s a huge part of it. It’s the difference between building a brand-new, state-of-the-art “smart home” and trying to renovate a 100-year-old house. In the new house, you can build all the latest technology right into the walls. In the old house, you’re constantly dealing with old wiring and crumbling foundations. The Gulf gets to build the new house.

But it’s not just about having a blank slate. It’s also about having leaders who are saying, “Let’s build the most incredible house the world has ever seen.” The lack of old infrastructure removes the obstacles, but it’s the bold vision that’s really driving the speed.

Do state-led digital agendas accelerate execution compared to private-sector-only ecosystems?

Yes, especially at the beginning. A government can do things that no single company can. It can align the entire country around a single goal. It can make the massive, long-term investments in things like 5G networks or data centers that are too risky for the private sector alone.

The best approach, though, is a partnership. The government builds the highway, and then it lets private companies build all the interesting businesses along the side of it. The government proides the foundation and the rules of the road, and the private sector brings the innovation and the speed. It’s that combination of state vision and private-sector energy that creates the fastest and most sustainable growth.

The Larger Question

Are we mistaking tool adoption for strategic intelligence?

Yes, all the time. It’s a huge mistake. It’s like buying a really expensive set of golf clubs and thinking that makes you a professional golfer. The clubs are just tools. The real skill is in how you swing, how you read the course, and how you handle the pressure. It takes years of practice to develop that.

It’s the same with AI. Buying a new AI software is easy. But building a truly intelligent company is hard. It’s about changing your culture so that everyone is constantly learning from data and making better decisions. The AI is just the tool that helps you do that. The real advantage comes from the wisdom you build as an organization. The tool is a commodity that anyone can buy; the wisdom is yours alone.

Ten years from now, what will distinguish organisations that thrived in the AI era from those that didn’t?

Ten years from now, the companies that thrived will be the ones that were the best at adapting. The world will be changing so fast that the ability to learn and change will be the most important skill of all. I think it will come down to three things:

• They will be learning machines. They will be obsessed with training and developing their people. They will have figured out how to make humans and AI work together as a seamless team, with each doing what it does best.
• They will be trusted. In a world full of deepfakes and misinformation, trust will be the most valuable brand asset. The companies that win will be the ones that are radically transparent about how they use data and AI. They will have earned the trust of their customers, and that will be a huge advantage.
• They will be deeply human. As technology gets more powerful, the best leaders will be the most human. They will be the ones who can inspire their teams, who can create a culture of creativity and belonging, and who can make work feel meaningful. In the age of artificial intelligence, real human connection will be the ultimate differentiator.

Rapid Fire

Efficiency or empathy? 

Empathy. Happy, engaged people are efficient people. Not the other way around.

Skill gap or system redesign? 

System redesign. A good system makes it easy for people to do the right thing. Don’t blame the person; fix the system.

AI as assistant or AI as operator? 

Assistant. It’s a tool to help smart people do their best work. The human should always be the one in charge.

Automation or augmentation? 

Augmentation. Automation is about cutting costs. Augmentation is about growing capabilities. I’m more interested in growth.

Speed or sustainability? 

Sustainable speed. Sprinting leads to burnout. You have to be able to run a marathon at a fast pace.

Replacement or reinvention? 

Reinvention. It’s a much more positive and powerful way to think about the future. It’s not about what we’re losing; it’s about what we can become.

Execution or experimentation? 

Experimentation. You have to constantly be trying new things to find out what works. The execution part is easy once you’ve found a winning idea.

Growth or survival? 

Growth. If you’re just trying to survive, you’re already losing. You have to be aiming for growth.

AI education or AI governance? 

Education. If people understand the technology, they will demand good governance. It has to start with knowledge.

One word that defines the future of work. 

Human. The more technology we have, the more we will value what makes us uniquely human.

---

---

Also Read: From Macron to Modi: How Leaders See the AI Job Shift in 2026

---