PETALING JAYA: Malaysia’s challenge in artificial intelligence (AI) is no longer about adoption, but turning that uptake into real economic and corporate value, says McKinsey & Co at a media roundtable held in conjunction with the release of “AI in South-East Asia: An Era of Opportunity” report.
McKinsey & Company Malaysia managing partner Vidhya Ganesan said an estimated two million small and medium enterprises (SMEs) have adopted AI, a recent case study showed.
However, only a small proportion of these SMEs have seen meaningful financial returns.
She said only a tenth of these businesses report value of any kind flowing through to the bottomline.
Vidhya said at the national level, it also does not lie in investment but how these investments translate into economic outcomes.
“In terms of raw investments into data centres, cloud, from the hyperscalers, there’s no shortage at all,” Ganesan said.
“But if you really look into the translation of that to how it is flowing through into value capture for gross domestic product, how many extra good jobs have we created? That is really where the tension is.”
She noted that Malaysia already has one of the highest concentrations of data centres globally, particularly in Johor, underscoring the scale of infrastructure investment taking place.
However, she stressed that the country must move beyond simply hosting such infrastructure and instead focus on capturing higher-value activities.
“It requires us to move up the value chain beyond being rent seekers just for our land, which is effectively going beyond innovation to starting to own intellectual property,” she said.
“It’s effectively ensuring we don’t repeat what we did with not being able to move up the semiconductor value chain.”
On the question of what Malaysia gains from hosting large-scale data centres, Vidhya Ganesan said while there were real benefits, there were also trade-offs.
“The disadvantages are very clear. The drag it creates on land, on water, energy is very clear and acknowledged,” she said.
However, she noted that the presence of hyperscalers also brings strategic spillover benefits beyond infrastructure investment.
“The fact that these hyperscalers have this quantum of investment in the country means the regional leadership, global leadership of some of these hyperscalers show up here every quarter,” Ganesan said.
She said these visits create direct engagement opportunities with local enterprises and government-linked companies, often leading to deeper collaboration.
Ganesan added that such interactions can create a “virtuous cycle” of investment, innovation and partnership, that would otherwise not materialise.
She also pointed to a potential “talent halo effect” from hyperscaler presence, provided Malaysia can fully capitalise on it.
She added that this would require supportive policies, particularly around attracting and retaining foreign talent, alongside ongoing domestic initiatives.
“Can we accelerate and do better? Absolutely yes,” she said.
The findings mirror broader regional trends highlighted in the McKinsey report, which found that while companies have invested in AI, financial impact remains limited.
“More than 70% of the companies are saying we are adopting AI, but we are not really seeing the transformative impact,” said McKinsey partner Vivek Lath, a principal author of the report.
The report surveyed 330 companies using AI across six South East Asian economies – Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam – panning multiple industries and company sizes.
Across the region, the report found that most organisations allocated between 11% and 40% of their technology budgets to AI, yet returns remain modest.
Roughly six in 10 companies reported less than a 5% improvement in earnings before interest and taxes, while nearly one in five saw no impact at all.
This pattern mirrors global trends, where many companies have seen some efficiency gains from individual AI use cases but struggle to scale that impact and translate it into bottom-line value.
This has created what Lath described as a paradox, where companies invests but struggle to extract value.
A key issue remains how companies that deploy AI do not fundamentally rethink how their businesses operate.
“If you are using AI without really re-imagining the processes, you are digitising the inefficiencies in the processes,” he said, adding that businesses need to have a “rewired” approach in which AI embeds into business-led transformation supported by changes in operating models, talent, and data capabilities.
Talent remains a key constraint in Malaysia, particularly the shortage of professionals who can bridge business and technology.
“The issue is much more one of what we call translators,” said Ganesan.
She added that value creation is disproportionately driven by mid-tier companies, suggesting policy and corporate focus should prioritise scaling these firms.
At the ecosystem level, Lath said governments have a key role in enabling broader participation through shared platforms and infrastructure.
“Governments should create platforms where some of the small scale players who could not have done things on their own are able to capture the benefit as well,” he said.
