Judgment Is the One Skill AI Cannot Replace

AI is changing more than how businesses work. It is changing how value is created, where competitive advantage comes from, and what leaders need to build for the future.

As AI capabilities become increasingly accessible, the technology itself will become less of a differentiator. The bigger question is what separates organizations that use AI to create lasting value from those simply adopting the same tools.

That shift extends beyond technology. It changes the role of human judgment, the talent organizations need, what enterprise buyers expect from their partners, and ultimately how leaders create value.

To explore what this shift means for business and leadership, Opus Technologies Chief Marketing Officer Nischala Murthy sat down with CEO Srini Rajamani.

His view is that AI is not simply another technology cycle. It is a broader economic and societal shift, and the organizations that thrive will be those that combine AI’s scale with human judgment, domain expertise, ownership and the ability to adapt.

Nischala: You’ve led through Y2K, the financial crisis, cloud computing, and now AI. What makes this AI wave fundamentally different?

Srini: Many things are different. For one, this shift has a remarkably visible starting point. ChatGPT brought generative AI into the mainstream almost overnight, and from there, adoption accelerated at a pace we haven’t seen with previous technology cycles.

But what makes AI fundamentally different is its reach. Previous waves primarily changed how businesses operated or how technology was delivered. AI is beginning to change the economics of entire industries, the nature of work, and even how value is created and distributed.

In the US, the AI-driven boom is already influencing equity markets, job categories and prices. At the same time, rising demand for memory and computing infrastructure is pushing up costs across the technology ecosystem. There are real positives and real negatives.

And the important point is that AI is no longer confined to the technology function or to productivity. It is touching almost every part of life and business.

Some jobs will be eliminated by AI. But history tells us that major disruptions also create new categories of work that were difficult to imagine beforehand. The same could happen again. The question is not simply which jobs disappear, but what new forms of work and value creation emerge.

We also don’t yet know how AI will reshape the distribution of opportunity. Will it widen the gap between the haves and have-nots? What guardrails will be necessary? And what will its impact be on children and other vulnerable parts of society?

Social media has already demonstrated how quickly technology can create unintended consequences. AI could do the same, but at a much greater scale.

So, yes, AI is enormously positive in many ways. But it is also a technology every leader needs to watch closely. The opportunity is significant; so is the responsibility that comes with it.

Nischala: You describe AI as a broader economic and societal shift. How is it changing the foundations of how value is created?

Srini: AI is not simply changing what we do with technology; it is changing the infrastructure, economics and human capabilities that sit underneath it. So, in my view, it is changing the foundations of how value is created.

Data centers now hold a significant share of the world’s digital knowledge, and the race is extending all the way down to power and infrastructure with companies even exploring data centers in space. AI could ultimately unlock access to vastly greater computing and energy resources than we have today.

And most people see only the visible layer, i.e., the chatbot or application. Underneath it are five interconnected layers:

1. Utility and power: the energy required to run AI at scale.
2. Chips: the compute that powers the models.
3. Data centers: where computing and storage happen.
4. The AI models themselves: from OpenAI and Anthropic to newer entrants such as DeepSeek and Kimi and,
5. Applications: the interfaces people interact with.

Most of the AI attention is on layer four. But disruption and opportunity are emerging across all five.

Nischala: With disruption and opportunity emerging across these layers, where will organizations find lasting competitive advantage?

Srini: Mindset is the starting point, but judgment is the differentiator.

Both of my children work in financial services, at companies that are fairly advanced technologically. When I talk with them, what strikes me is that the real capability leaders need to develop is the ability to make decisions at the edge.

AI can hallucinate. It can give you information that is wrong. You must challenge it, validate it and decide when to trust it.

There is a simple analogy. Before GPS, people learned to read maps. Today, many people struggle to navigate when GPS is unavailable. The technology solved a problem, but we also outsourced a capability.

The same risk exists with AI. If it writes our emails, summarizes our documents and increasingly makes recommendations for us, we cannot allow our own judgment to atrophy.

The organizations that win will not be those that simply use AI more. They will be those that combine AI’s scale with human judgment.

Nischala: We’re seeing growing demand for people who can apply AI to real-world business problems. OpenAI’s recent acquisition of Northslope seems to reflect that shift. What does it tell us about the kind of talent the market will value?

Srini: It tells me that the market is placing a premium on people who can bridge technology and business, not simply execute technical tasks.

OpenAI’s Deployment Company acquired Northslope, an applied AI firm founded by former Palantir engineers, to expand its ability to embed forward-deployed engineers with customers and move AI from models into real-world operations.

That is an important signal. Large service companies will continue to have a market for lower-end work. But as automation accelerates, that market will face increasing pressure. The premium will shift toward people who can understand the technology, the industry, the client and the business problem and bring those pieces together.

This is very different from the traditional model of breaking work into smaller tasks and assigning each task to a specialist.

The emerging talent model is a technologist who understands the client’s industry, can manage stakeholders, and can sit across the table from a CXO and co-create the solution.

That is where I believe value will increasingly concentrate.

Nischala: What does this changing talent model mean for what enterprise buyers expect from engineering partners?

Srini: It means the traditional input-based model is becoming less relevant.

For years, the industry sold people, skills and billable hours. Customers are increasingly asking a different question: what business outcome will you deliver, and how quickly?

Take a supply chain platform. The question is no longer how many developers it takes to build it. The question is: if it takes five days to ship a product today, can we reduce that to five hours?

Answering that requires a systems thinker, domain expert, engineer, tester and operations expert working together toward one outcome.

Customers are no longer saying, “I need 200 developers.” They are saying, “Here is the outcome I need. You figure out the best way to deliver it.”

It is like ordering a ride when you have a flight to catch. You don’t care how many drivers are available or how many cars the company owns. You care that a car arrives in five minutes. You are paying for the outcome – not the inputs.

That is the fundamental shift, from selling capacity to creating value.

Nischala: As enterprise buyers focus more on outcomes and creating value, what capabilities matter most? Does deep industry expertise become a differentiator?

Srini: Yes. But domain expertise alone is not enough. You need domain expertise and the mindset to take ownership of the outcome.

Domain is about understanding an industry deeply. Mindset is about seeing a problem, taking responsibility for solving it, and using whatever technology is available to get there.

A lot of people are used to being told what to do, when to do it and why. That model is changing.

The people who succeed will be able to look at a client and say, “I understand your problem. I will assemble the right capabilities and deliver the outcome.”

That combination of domain knowledge, technology fluency, ownership and judgment is difficult to automate, and even harder to teach through a traditional curriculum.

Nischala: We’ve talked about how work and value are changing. What does that require of leaders themselves?

Srini: Get comfortable with change.

People who expect the world to behave as it did in the past will be disappointed. Change that once took years can now happen in weeks or even days. The AI landscape itself is evidence of that, from OpenAI and Anthropic to lower-cost entrants such as DeepSeek and Kimi.

But being the latest is not the point. Being able to adapt to what is next is.

The leadership advantage will come from staying curious, staying humble and continuously learning.

I was recently with a client, and they stay humble and curious even though they have a market value of $100 billion! That is the culture we are trying to build at Opus because in a world where machines can increasingly provide the answers, the leaders who keep asking better questions may have the greatest advantage.