AI Is Making Everything Faster. I Chose to Slow Down.

After 15 years of working, I made a decision that probably wasn't the most efficient one.

When my family moved to Singapore, I could have started looking for my next job.

I didn't.

Instead, I went back to university and started a Master's in Economics at Nanyang Technological University.

Then I made an even less efficient decision.

My master's was originally a one-year programme. But I competed for an exchange opportunity at Waseda University in Japan — knowing that if I took it, my one-year master's journey would stretch to almost two years.

I went for it anyway.

During this period, I also turned down several opportunities to return to work.

From a conventional career perspective, none of this looks particularly optimized.

And we are living in an age obsessed with optimization.

AI writes faster. Codes faster. Searches faster. Analyses faster.

Companies want shorter cycles. Professionals want faster upskilling. Careers are increasingly discussed in terms of acceleration.

So why did I deliberately slow mine down?

Experience Can Become Path Dependence

Fifteen years of professional experience gives you something valuable: pattern recognition.

You have seen organizations grow and restructure. You have watched people succeed and fail. You have developed instincts about talent, leadership, incentives, culture, and how organizations actually work beyond what appears on an org chart.

But experience has another side.

It can become path dependence.

The longer we work, the easier it becomes to interpret new problems through frameworks that worked for us in the past.

And AI made me increasingly uncomfortable with some of mine.

What exactly is a "job" when AI can perform a growing share of the tasks inside it?

What does expertise mean when knowledge that once took years to accumulate can be accessed in seconds?

What should managers manage when AI begins to perform parts of cognitive work?

What happens to experienced white-collar professionals when some of the capabilities that made them valuable become partially automatable?

And when a company introduces AI, is it actually transforming work — or simply adding another tool on top of an organizational structure designed for another era?

I realized that before deciding what I wanted to do next, I needed to update the operating system I was using to understand work itself.

That was one of the reasons I chose NTU.

Not because I wanted another line on my CV.

I wanted distance from the assumptions I had accumulated over 15 years.

Economics gave me a different lens for thinking about productivity, labor markets, incentives, technology, and human capital.

And building with AI alongside my studies made the questions even more real.

The more I used AI, the less interested I became in the simple question:

"Will AI replace jobs?"

I think the more interesting question is:

What happens to the tasks, organizations, and people around AI when cognitive work begins to change?

Then Came Waseda

When the opportunity to compete for an exchange at Waseda University came up, there was an obvious reason not to apply.

Time.

Going to Japan would extend what could have been a one-year master's into a much longer journey.

For a mid-career professional, the opportunity cost is real.

Every additional semester outside the workforce can be calculated in salary, seniority, network momentum, and missed opportunities.

And I did calculate it.

I still applied.

Because by then, I had started to think differently about what I was optimizing for.

I wasn't trying to collect a degree as quickly as possible.

I was trying to use this period to answer questions that may shape the next 10 or 15 years of my career.

Japan made those questions even more interesting.

This is what I will be researching at Waseda:

From Displacement to Augmentation: AI-Driven Task Reallocation and Human Capital Transformation in Japan and Singapore.

At the center of the research is a question I have become increasingly obsessed with:

When AI takes over parts of cognitive work, how do organizations reallocate the work that remains?

I want to understand which white-collar tasks are actually being automated, which new human-centric tasks are emerging, and why organizations adapt differently even when they have access to similar technologies.

Japan and Singapore offer a fascinating comparison.

Japan combines labor shortages and an aging population with relatively strong employment stability and deeply embedded organizational practices.

Singapore has taken a different path, with a more flexible labor market and a strong national emphasis on continuous reskilling through initiatives such as SkillsFuture.

The technology may be similar.

The organizational response may not be.

And I am particularly interested in what happens to workers aged 40 and above.

Mid-career upskilling is different

When we talk about upskilling, we often imagine learning something new.

Learn AI.

Learn coding.

Learn data.

Take a course. Earn a certificate. Add another capability to the CV.

But I increasingly think mid-career upskilling is harder than that.

Because after 15 years, the challenge isn't simply adding knowledge.

It is deciding what to **unlearn**.

Which assumptions that helped me succeed in the past are still valid?

Which ones belong to an organizational world that is disappearing?

Which parts of my experience can AI amplify?

Which parts are becoming less valuable?

And what new capabilities become more important precisely because machines are becoming better at cognitive work?

These are not abstract research questions for me.

In some ways, I realized I wasn't only researching this transition.

I was living through it myself.

I am a mid-career professional studying what happens to mid-career professionals when technology changes the value of human capital.

That realization changed how I think about the opportunity cost of these two years.

Maybe Career ROI Isn't Speed

I could have finished faster.

I could have returned to the job market earlier.

Those would have been perfectly rational choices.

But there are moments in a career when optimizing for speed may optimize the wrong variable.

AI is dramatically reducing the time required to produce, search, analyze, build, and learn.

That makes speed incredibly valuable.

But perhaps it also makes something else more valuable:

knowing where you are going before accelerating.

For me, these two years are not a pause from my career.

They are part of it.

A chance to combine 15 years of organizational experience with economics, AI, research, building, and a comparative perspective across Singapore and Japan — and to see which parts of my own thinking survive the collision.

I don't know exactly where that will lead yet.

And strangely, I am more comfortable saying that today than I would have been a few years ago.

Because sometimes upskilling isn't about moving faster.

Sometimes it is about creating enough distance to see what you need to become next.

In a world obsessed with accelerating everything, perhaps the real advantage is knowing what deserves to be accelerated — and what deserves more time.