Malaysia AI adoption surges 35%, yet most fail to transform

• Malaysia AI adoption surged 35% to reach 2.4 million businesses.
• 73% remain stuck at basic efficiency levels, not driving innovation.
• “Two-tier AI economy” emerges: startups outpace large enterprises in advanced AI integration.

Malaysia’s AI adoption has reached a important inflection point – but not necessarily the one policymakers might have hoped for. While headline numbers show impressive growth, with over 2.4 million businesses now using artificial intelligence technologies, new research from Amazon Web Services reveals an uncomfortable truth: the vast majority are barely scratching the surface of what AI can do.

The AWS-commissioned studysurveying 1,000 Malaysian businesses representative by size, sector, and region, found that 27% of companies have adopted AI – up from 20% last year, representing a 35% growth rate. This translates to 630,000 new businesses implementing AI technologies during 2024 alone, or more than one new adoption every minute throughout the year.

But dig beneath these encouraging statistics, and a more complex picture emerges.

The surface-level AI economy

According to the report, 73% of Malaysian businesses using AI remain focused primarily on basic applications – scheduling assistants, ready-made data analysis tools or off-the-shelf cybersecurity solutions. Companies are capturing incremental efficiency gains rather than the transformative innovation that AI promises.

Only 10% of AI-adopting businesses use the technology to a significant degree. “While AI is increasingly being widely used in businesses in Malaysia, a small segment of organisations has already harnessed the technology’s most transformative potential,” the report states. “Most companies are still exploring AI’s benefits, missing out on the deeper strategic advantages it can offer.”

The retail sector exemplifies the trend, with 85% concentrated at the most basic stage of AI adoption – the highest proportion of any industry surveyed.

The two-tier divide: Startups vs. enterprises

Among startups – defined as businesses founded in the last two years that offer new products or innovations – 48% are using AI in their operations. Crucially, only 46% of these startups remain at basic AI adoption levels, compared with 74% of large enterprises (those with 500+ employees).

More tellingly, 31% of startups are developing new AI-driven products and services, while only 15% of large enterprises are doing the same. Meanwhile, 27% of startups have placed AI at the core of their business proposition, and 26% are using AI for its most advanced uses – figures that dwarf large enterprise performance.

“The ‘two-tier’ AI economy – with tech-driven startups outpacing larger, established enterprises in AI innovation – can impact Malaysia’s AI-driven growth and innovation in the years to come,” the report warns.

The benefits are real – for those who commit

The economic case for deeper AI integration is compelling. Among businesses that have adopted AI, 65% report revenue increases averaging 19%, while 72% say they’ve seen significant productivity improvements.

The research shows that 74% of AI adopters believe the technology will likely increase their growth in the next year, and 67% expect cost savings averaging 15%. The gains are enabling businesses to redirect focus toward enhancing customer service (38%), investing in employee training (33%), and developing new products and services (28%).

Yet 31% of businesses say they need a clearer understanding of AI’s return on investment – despite the proven results from their peers. The knowledge-action gap suggests a communication challenge as much as a technological one.

Three barriers

The research identifies three important obstacles preventing Malaysia’s AI adoption from deepening beyond surface-level applications:

Skills shortage tops the list. With 52% of businesses citing a lack of digital skills as their primary barrier, the talent gap is acute. Companies report that adapting to new digital technologies (43%), data analysis and interpretation (39%), and the basics of AI and machine learning (32%) are the most lacking skills in their workforce. Businesses say they would increase salary offers by 34% to candidates with strong AI skills – a premium that reflects desperation as much as opportunity.

Only 22% of employees have participated in digital training or upskilling in the past year, while businesses expect AI literacy to be important for 54% of jobs in three years. Just 29% of companies feel prepared with their current skillset.

Regulatory uncertainty looms large, even in Malaysia’s currently light-touch environment. While the country lacks AI-specific regulation, 49% of businesses hope future rules will provide a stable regulatory framework. However, concerns about increased compliance costs (42%), slowed innovation (41%), and lack of legal certainty around AI use (40%) weigh heavily on decision-makers watching global regulatory developments.

Perceived costs create hesitation, with 39% citing upfront expenses as a key barrier – despite the documented ROI from early adopters.

Sector-specific momentum

Some industries are pulling ahead. Technology and professional services lead adoption at 49%, followed by financial services (42%) and manufacturing (39%). Financial services also dominate advanced AI integration, with 21% at the most sophisticated stage, followed by technology (20%) and healthcare (15%).

Notably, 81% of Malaysian businesses overall – and 83% of startups – believe AI will transform their industry in five years, suggesting awareness isn’t the problem.

The path forward

The research concludes with three recommendations that extend beyond the usual calls for digital transformation:

First, accelerate private sector digital adoption through targeted skills programmes. The government’s RakyatDigital initiative, which reached over 1 million users in just six months since January 2024, demonstrates demand. But more industry-specific programmes developed through public-private-academic collaboration are needed.

Second, establish pro-innovation regulatory frameworks that focus oversight where potential harm is greatest while allowing low-risk applications to scale quickly. The report advocates for AI regulatory sandboxes where companies can test solutions under supervised conditions.

Third, increase public sector AI adoption. With 71% of Malaysian businesses saying they’re more likely to adopt and expand AI use when the government leads by example, and 76% of startups calling public sector adoption “crucial” to their scaling ability, the government’s own digital transformation becomes a market enabler.

The AWS study, based on surveys of 1,000 businesses and 1,000 members of the public (both representative samples), paints Malaysia as a country at a crossroads. The foundation is solid: 2.4 million businesses using AI, government programmes gaining traction, and strong momentum in the finance, technology, and manufacturing sectors.

But the gap between early adoption and transformative integration threatens to crystallise into a permanent divide – one where Malaysia’s most agile companies surge ahead while its largest enterprises, and the broader economy, struggle to keep pace. How quickly that gap closes may determine whether Malaysia becomes a regional AI leader or a cautionary tale of squandered potential.

Want to experience the full spectrum of enterprise technology innovation? Join TechEx in Amsterdam, California, and London. Covering AI, Big Data, Cyber Security, IoT, Digital Transformation, Intelligent Automation, Edge Computing, and Data Centres, TechEx brings together global leaders to share real-world use cases and in-depth insights. Click here for more information.