Malaysia Lost 31,000 Jobs in 4 Months.
AI Wasn't the Cause.
That's What Should Scare You.

The numbers

Malaysia's Economy Minister confirmed 7,057 workers lost their jobs in April 2026 — a 21% jump from March, reversing what had looked like a declining trend. The total for the year so far stands at 31,082. That is not a rounding error. That is more than 31,000 people who went to work, and then did not.

The numbers were concentrated in manufacturing and export-driven industries. Selangor and Kuala Lumpur together accounted for more than half of all layoffs. Penang and Johor — heavily dependent on E&E and cross-border trade — remain the most exposed outside the Klang Valley.

Most people saw this headline and moved on. Economic cycle. Trade uncertainty. Global conditions. Nothing to do with me.

I think that response is exactly wrong.

This wasn't AI

Let me be precise about the cause. The majority of Malaysia's 2026 retrenchments are being driven by manufacturing restructuring — companies recalibrating production capacity in response to US tariff uncertainty, the shifting US-China trade dynamic, and a global slowdown in the electronics demand cycle. The E&E sector alone accounts for nearly 40% of Malaysia's total exports, and when global tech demand softens or supply chains re-route, Malaysia feels it faster than most.

This is not an AI story. It is a trade story. Companies are not replacing these workers with AI models. They are cutting headcount because external demand dropped and the economics of keeping full capacity no longer work.

And that is the part that concerns me most.

AI is still loading

Globally, the picture looks different. In the United States, a Challenger report found that AI was cited as the cause of 26% of all job cuts in April 2026. Eighty thousand workers have been displaced by AI-driven automation in 2026 alone — across companies like Accenture, Amazon, Citigroup, Intel, and Microsoft. These are not factory workers. These are analysts, coordinators, support roles, junior executives — the exact categories that make up the middle layer of most Malaysian workplaces.

The Harvard Business Review published research showing that companies are cutting these jobs not because AI has already delivered on its promise — but because of AI's potential. Businesses are making workforce decisions based on what they believe AI will be able to do, not what it demonstrably does today. The jobs are gone regardless.

In Malaysia, this wave has not yet arrived at scale. The companies making AI-driven workforce cuts right now are primarily US and European multinationals. But many of them operate in Malaysia. Their local headcount decisions will reflect their global workforce strategies, with a lag of twelve to twenty-four months. What is happening in their US and European offices today is a reasonable preview of what arrives here.

What I see every day

I build AI infrastructure. It is my job to think about where AI is going, what it will automate, and what it will leave behind. And I find myself increasingly unsettled by the gap between what I know about this technology and how most people around me are living.

Every day I see young Malaysians — educated, capable, ambitious — going about their lives without any apparent sense that the career they are building might look fundamentally different in five years. The weekend plans. The lifestyle content. The conversations about what car to buy next. Not wrong. Not shallow, necessarily. Just completely disconnected from what is happening in the labour market and what is coming next.

AI, to most of them, is a productivity tool. Something that helps you write faster, summarise documents, generate images. Useful. A bit interesting. Not urgent.

I do not think that is right. I think AI is the most significant restructuring of the labour market in a generation — and it is arriving on a timeline that most people are not taking seriously.

The real question

Here is what I keep asking myself: if Malaysia's job market is already absorbing 31,000 retrenchments in four months — from causes that have nothing to do with AI — what does the labour market look like when the AI displacement wave actually arrives at scale?

Manufacturing jobs are being cut because of trade dynamics. The companies cutting them are not replacing the workers — they are simply removing the capacity. When AI automation starts eliminating knowledge work — the analyst role, the coordinator role, the junior account manager, the customer service team — the same math applies. The work still gets done. By fewer people. Or by no people.

Malaysia's workforce is not prepared for this. Not because people are not capable — they are. But because most of the preparation being offered is at the surface level. Learn to use AI tools. Take a prompt engineering course. Add ChatGPT to your CV. These are not wrong suggestions. But they address the tool, not the transformation. The workers who survive and grow through this period are not the ones who learned to use AI. They are the ones who understood what AI changes about their industry — and repositioned themselves ahead of that change, not after it.

I am not writing this to cause panic. I am writing it because I think the gap between what is actually happening and how most people are responding to it is too wide to stay quiet about.

On my end, I am meeting with one of Malaysia's top private universities to discuss a new generation of short courses — not the generic "Introduction to AI" kind, but focused, practical programmes built around what the market actually needs right now. Short, effective, and designed for the AI era rather than retrofitted from old curricula. I am looking forward to that conversation. It is one of the things that needs to happen.

The numbers are already there. 31,082 jobs. 47% increase. And the wave that will actually reshape the labour market has not yet arrived.

Now or Never.

Published by IMA AI — June 2026. Retrenchment data: PERKESO / Ministry of Economy Malaysia / SAYS.com.