Themes from “Competing in the Age of AI”: A Reflective Walk Through Strategy, Leadership, Ethics, and More

While researching how AI is redrawing the map of international business, I picked up Competing in the Age of AI by Marco Iansiti and Karim R. Lakhani—driven by both curiosity and urgency. I wanted to understand how AI transforms technology, how it changes the way companies are structured, how they lead, and how they stay accountable in a fast-changing world.

As I read, I started grouping insights not just by chapters but by themes that felt deeply relevant to the work many of us are trying to do—whether you’re building a tech startup, leading a traditional firm through change, or simply trying to understand the shifting landscape. Here’s my take, organised by four key themes: Leadership, Strategy, Ethics, and Architecture—with a few stories along the way.

About the Author

Marco Iansiti and Karim R. Lakhani are professors at Harvard Business School. Iansiti specialises in technology strategy and the digital transformation of firms. Lakhani is a leading researcher on artificial intelligence, open innovation, and platforms. Together, they’ve advised global firms and led executive programs on how AI is reshaping business and competition.

Key Takeaways

• Leadership Must Evolve: The best leaders in the AI era reimagine culture, structure, and decision-making—not just tech adoption.
• Strategy is Systemic: Advantage now comes from platform orchestration, data flows, and feedback loops—not static positioning.
• Ethics Scales Too: AI amplifies bias and surveillance if left unchecked. Responsible design is non-negotiable.
• Rearchitect the Firm: Winning firms run AI factories. It’s not about overlaying AI—it’s about rebuilding the operating model from the core.

Leadership: Adoption and Reinvention

Leadership in the age of AI is about fundamentally rethinking how decisions are made, how teams are structured, and how accountability is built into the very fabric of the organisation.

Chapter 5, Becoming an AI Company, dives into this challenge headfirst. Microsoft is the standout case here. Under Satya Nadella, Microsoft didn’t just become more digital—it became a learning organisation. It shifted from a product mindset to a platform mindset, from silos to cross-functional teams, from managing products to orchestrating ecosystems. That required leadership with both humility and boldness.

Later, in Chapter 10, the authors bring this full circle: AI is a technological shift as well as a leadership test. The best leaders build ethical governance into AI strategy, empower frontline teams to experiment, and rethink what it means to “scale” in a world where software—not people—is doing most of the operational heavy lifting.

Three Key Examples:

• Microsoft: Nadella’s cultural reset emphasised empathy, learning, and AI as a core driver—not an add-on. “AI is the runtime that is going to shape all of what we do,” as Nadella puts it.
• Fidelity Investments: Built an internal AI Center of Excellence and democratised data literacy across the organisation. Leadership backed it with both training and structural change.
• Ping An: Transformed from a traditional financial services firm to a tech-first platform spanning health and finance—driven by visionary leadership willing to challenge internal orthodoxy.

Strategy: From Moats to Flywheels

Traditional strategy is all about positioning. But in the AI era, as laid out in Chapter 6, strategy is more about participation—being part of the right data flows, networks, and feedback loops.

Three things now matter more than anything:

• Your network position (how well-connected your users, suppliers, and partners are),
• Your data advantage (not just data collection, but the ability to act on it in real time),
• And your ability to build a platform that lets others create value too.

Uber is a great example from this chapter. It’s not just a taxi alternative—it’s a matchmaking engine fuelled by real-time data. Every interaction sharpens its algorithms, making it harder for slower-moving competitors to keep up.

And in Chapter 7, we see what happens when these fast, AI-native firms run into traditional businesses. Spoiler: it’s not pretty. Airbnb didn’t beat hotels on price or comfort—it won on data and network effects. Tesla didn’t beat car companies with better factories—it beat them with better feedback systems and software.

Three Key Examples:

• Uber: A dynamic pricing and matching engine that grows smarter with every ride. Its strategy isn’t just about scale—it’s about learning.
• Airbnb: Turned lodging into a platform play—using trust mechanisms, user reviews, and AI-powered recommendations to build a global marketplace.
• Tesla: Every mile driven feeds into its neural network, improving self-driving capability. The firm learns faster than incumbents can redesign.

“Strategic moats are now dynamic—created and lost faster than ever before.”

Ethics: The New Cost of Scale

Chapter 8 made me pause. We often celebrate scale in digital business, but what if the very things that make AI powerful also make it dangerous?

AI systems can amplify bias, deepen surveillance, and accelerate misinformation. And because they scale so fast, the consequences are often outsized and invisible. The chapter walks through tough examples—from facial recognition bias to targeted ad systems that unintentionally reinforce discrimination.

The takeaway? Ethical frameworks can’t be afterthoughts. They need to be part of the operating system. Leaders must set the tone, yes, but organisations also need independent auditability, explainability, and internal governance teams with teeth.

Chapter 10 picks this up again, arguing that responsible AI is not a “compliance task” but a leadership mandate. That shift in framing really stayed with me.

Three Key Examples:

• Facebook: Algorithms optimised for engagement led to misinformation propagation. Highlighted the risks of AI at scale without proper ethical guardrails.
• Google Ads: Showed how ad delivery could unintentionally vary by race or gender—revealing latent biases in training data and model outputs.
• Predictive Policing Tools: Reinforced systemic bias by learning from historically biased datasets—raising urgent questions about accountability.

“Algorithms are not neutral. They reflect the data they are trained on—and the intentions of those who deploy them.”

Architecture: Rearchitect, Don’t Just Digitise

Framework: Human-at-the-Edge Operating Model

Early on—in Chapters 2 through 4—the book makes one thing clear: digitising your current operations is not the same as becoming an AI-driven firm.

What’s needed is a full rearchitecture of the firm. The standout metaphor is the “AI factory” from Chapter 3—a system that continuously converts data into improved predictions, decisions, and actions. Netflix embodies this beautifully. Every user interaction feeds a loop of better recommendations, smarter content acquisition, and more engagement.

Then in Chapter 4, we see Amazon as a masterclass in modular architecture. Every internal service was rebuilt to be API-accessible. That means teams could innovate independently, share data efficiently, and integrate AI seamlessly into the logistics chain. It’s not about putting AI on top of your current processes—it’s about rebuilding your processes so that AI is the core engine.

Three Key Examples:

• Netflix: Built an AI factory around personalisation. Their A/B testing culture and recommendation system are deeply embedded in the operating core.
• Amazon: Created internal modular services with APIs, enabling every business unit to innovate, learn, and scale AI features independently.
• Ocado: Transitioned from an online retailer to a tech platform providing AI-driven fulfilment systems to global partners.

“AI doesn’t just support decisions—it becomes the system that makes them.”

Concept	Netflix	Amazon	Ping An
AI Factory Use	Content recommendation	Supply chain, pricing, CX	Insurance, health, finance
Data Utilization	Viewing behavior, metadata	Shopping history, device, location	Claims, health data, financial transactions
Rearchitected System	Experimentation and personalization engine	Modular APIs, robotics, cloud-native ops	Integrated platform across services
Outcome	Scalable global personalization	Hyper-efficient operations, service scope	Full-service digital ecosystem

Closing Thoughts

Reading this book, I was reminded that AI isn’t just a technological disruption—it’s an organisational one. The firms that will thrive aren’t just the ones that use AI. They’re the ones that think like AI: fast, experimental, ethical, and deeply integrated.

The challenge ahead isn’t only to adopt AI tools. It’s to reimagine our firms, our leadership styles, our strategic assumptions, and our ethical guardrails. And maybe, if we do it right, we won’t just compete in the age of AI—we’ll help define it.