Opinion
Make The Edge Malaysia your preferred
Malaysia offers a revealing view of the artificial intelligence (AI) productivity paradox. Nearly one in four workers already qualifies as a “Frontier Professional”, someone who uses advanced AI tools to reshape daily work, compared with 16% globally. Seven in ten Malaysian AI users also say they now produce work they could not have created a year ago, according to Microsoft’s 2026 Work Trend Index. Capability is rising quickly. A key question needs to be what this new capacity is producing beyond output. Are people better off? Is work more meaningful? Are communities and the environment benefitting? Our fixation on productivity above all else rests on five myths. Four are old. Generative AI gives them new speed and scale.
The first myth is that growth wears the mask of progress. Productivity and progress are often treated as the same thing: more output means more growth, and more growth means more progress. Productivity measures what an hour produces; progress asks what that hour made possible, and what it cost the person producing it. Malaysia’s labour productivity per hour rose 3.8 % to RM45.10 in the third quarter of 2025, according to Department of Statistics data. On paper that looks like progress. But according to the EY 2025 Work Reimagined Survey, 68 % of Malaysian employees said their workload had increased, even though 81 % of generative AI users reported significant time savings. If AI is genuinely saving people time, that time is not showing up as rest or relief; it is being absorbed by more output. Productivity went up but nothing about the experience of work got easier. This gap, between an efficient hour and a bearable one, is where it should be possible to measure progress but where, according to the survey data, it is currently missing.
The second myth is the tendency to count what is easy to count. Underneath most performance systems sits an unspoken assumption: that what gets measured is what matters most. Dashboards reward on what they can crunch. Emails cleared, reports filed, meetings attended and response times are all easily captured through a performance system, so they quietly become the definition of good work, whether or not they are. Deloitte’s 2025 Global Human Capital Trends found respondents spending 41% of each day on work they said did not contribute to organisational value. That means nearly half the working day was going towards things the dashboard would happily record, while the people doing them consider them not to be adding value. Goodhart’s Law describes a familiar trap: once a measure becomes a target, people start optimising for the measure. A spotless inbox may signal discipline, avoidance, or both. The more important question is simple: did the work improve the outcome?
The third myth is that we borrow worth
The third myth is that we borrow worth from output. Ask a colleague how the week went and notice how quickly the answer becomes a ledger: finished, delivered, cleared, hit, exceeded. Underneath that habit there is often a belief that we rarely consider: that a person’s worth depends on what they produce. Psychologists call part of this contingent self-worth: our sense of value becomes tied to achievement, and research links such patterns to greater stress and costs to learning and well-being, as shown in self-worth research. Jobstreet’s 2026 Workplace Happiness Index suggests that this belief may not be paying off. 70% of Malaysian workers say they are happy at work, yet only one in three are satisfied with their stress levels and 41% feeling burnt out or exhausted. People are producing and seem to be telling themselves that production should feel like enough; it is not. Even taking a break can feel inappropriate unless it is renamed “recovery” so that it still sounds useful.
The fourth myth is that we slip into a world where “doing” is recognised over “mattering”. Work can be completed without becoming meaningful. Meaning often grows from contribution, connection and craft: who benefitted, why the task mattered and whether human judgment improved the result. Volume metrics capture little of this. A teacher can grade 80 scripts by the deadline and remember no student. A manager can leave six meetings with a full calendar and an empty decision log. When “doing” becomes the unit of value, “mattering” disappears from the dashboard first, and from the day soon after.
The final myth is the belief that speed is simply what AI delivers; that faster output means faster progress. The evidence suggests something closer to the opposite. AI can make people feel faster without making them faster, and even where it does save real time, that time rarely stays saved. The perception gap matters too. In a randomised controlled trial, 16 experienced open-source developers working on familiar codebases expected AI to cut task time by 24%. With AI, they took 19% longer. Afterward, they still believed it had made them roughly 20% faster, according to the METR trial. The study is narrow, yet the lesson is clear: felt speed can differ from measured speed. A related BetterUp Labs and Stanford study found 41% of workers had encountered “workslop”, polished-looking AI output that forces someone else to repair it, with nearly two hours of rework per instance.
Where AI does save genuine time, that saving
And where AI does save genuine time, that saving rarely becomes rest. Malaysia’s EY findings capture the tension: 93% of employees use GenAI at work; 81% of users report significant time savings; 76% report improved work performance; yet 68% say workloads have increased regardless. Time saved on one task can be filled by more tasks, higher expectations, reviewing work and coordination. Efficiency becomes capacity, and capacity attracts demand.
In Malaysia, adoption is also moving faster than organisational readiness. ManpowerGroup’s Talent Barometer reports 60% of Malaysian workers now use AI in their roles, up seven points from 2025, while confidence in using technology has weakened as training and job-security concerns persist. Microsoft’s 2026 findings add another warning: only 32% of Malaysian AI users say leadership is clearly and consistently aligned on AI.
We need to move beyond simply counting and work out how we navigate a practical return to purpose. A wider definition of ambition can loosen these myths. It starts and ends with two questions:
- Ask yourself at the beginning of the day: how can I use today to do something that matters to me and reflects the values I want to live by?
- Ask yourself at the close of each day: what became better because of how I spent my day; for me, for someone else, or for the world around me?
Meaning rarely appears on a productivity dashboard
Meaning rarely appears on a productivity dashboard. Make room for it anyway. Put a metric-free “me-moment” in your calendar: time to dream, explore, think and wander off-screen. Your life is larger than the sum, and speed, of your outputs.
Dr Cornelia C Walther is associate professor at Sunway University’s Institute for Global Strategy and Competitiveness (IGSC) and senior fellow at the Sunway Centre for Planetary Health (SCPH). She is a senior fellow at Harvard, the Wharton School and an external adviser on Hybrid Intelligence at UNFPA.
