The International Labour Organisation has released a comprehensive assessment of artificial intelligence's potential impact on Southeast Asian labour markets, revealing that generative AI stands poised to influence the working lives of approximately 80 million people across the ASEAN region. Despite this substantial figure, the organisation emphasises that the threat of widespread employment disruption remains limited at present, offering a more nuanced picture than the apocalyptic predictions sometimes heard in public discourse about automation and job displacement.

According to ILO projections for 2025, just under one-quarter of ASEAN's total workforce—specifically 22.9 per cent, or nearly 80 million workers—operates within occupations that face more than minimal exposure to generative AI technologies. This exposure assessment considers both the structural characteristics of different job categories and actual patterns of AI deployment emerging across the region. The finding suggests that while the potential reach of AI is remarkably broad, the distribution of risk is highly uneven, with concentration in particular sectors and skill levels.

The study introduces critical nuance by distinguishing between exposure levels. While nearly 80 million workers experience some degree of exposure to generative AI, only 3.3 per cent of the workforce—representing approximately 11.7 million people—occupy roles classified as having the highest exposure category. This distinction is crucial for policymakers and workers alike, as it suggests that fears of mass displacement may be overstated when one examines where AI is actually concentrated. Conversely, roughly two-thirds of ASEAN employment remains concentrated in occupations with no identified exposure to generative AI, indicating that vast swathes of the regional economy operate in sectors where automation remains largely irrelevant.

Geographic variation in AI exposure across ASEAN is striking and reveals significant differences in economic structure and digitalisation levels. Singapore leads the region decisively, with 42.2 per cent of its workforce operating in occupations with more than minimal AI exposure—a figure that reflects the city-state's position as a global financial and technology hub with advanced digital infrastructure. The Philippines follows at 28.1 per cent, a ranking that underscores its growing service and information technology sector, while Indonesia, Vietnam, and Thailand cluster between 20 and 22 per cent. This divergence highlights how AI's labour market impact will not be uniform across ASEAN, but instead concentrated initially in more developed and service-oriented economies.

Despite the broad exposure, evidence of large-scale job displacement has yet to materialise across the region. The report notes that employment in occupations with high AI exposure has actually continued to expand, suggesting that rather than simple job elimination, the region may be experiencing a transformation in job quality and content. Generative AI adoption itself remains in early stages, with uptake concentrated heavily in technology-intensive sectors and roles, while white-collar administrative and office work—categories theoretically vulnerable to automation—have seen comparatively limited AI integration so far. This gap between exposure and actual implementation suggests that organisational adoption barriers, cost considerations, and regulatory uncertainty may be slowing the pace of disruption.

A particularly significant finding involves the gender dimension of AI exposure, which reveals a surprising paradox. Women are more than twice as likely as men to work in occupations with high generative AI exposure, primarily because they dominate clerical, administrative, and certain professional roles that AI systems are designed to augment or replace. This concentration reflects broader labour market segregation patterns across Southeast Asia, where women are overrepresented in routine office and administrative work. The implications are profound: without deliberate intervention, AI-driven automation could disproportionately affect female employment in the region, potentially widening existing gender wage gaps and reducing women's economic participation.

Young workers between ages 15 and 24 exhibit exposure levels broadly similar to adult populations, contradicting assumptions that generational differences in digital literacy necessarily translate to differential vulnerability to AI. This finding suggests that age-based protections or training strategies may need reconsideration, with attention instead focused on occupational category and educational attainment as more reliable predictors of exposure. The convergence of youth and adult exposure levels also implies that entry-level workers entering ASEAN labour markets now will need to develop AI-adjacent skills regardless of age, necessitating fundamental shifts in education and training systems across the region.

Singapore emerges from the ILO analysis not only as the region's leader in AI exposure but also in preparedness, boasting a globally competitive ecosystem characterised by advanced digital infrastructure, deep talent pools, and a coordinated whole-of-government implementation strategy. This readiness gap is perhaps more consequential than exposure disparities themselves, as it will determine which countries and workers can benefit from AI-driven productivity gains versus those left behind. Malaysia, Thailand, and Vietnam face particular challenges in closing this preparedness gap, requiring significant investments in both technical infrastructure and human capital to ensure their workforces can adapt to AI-augmented roles.

The ILO identifies a critical preparedness gap across ASEAN that will shape how the region navigates AI's labour market implications. The organisation outlines four priority areas: establishing human-centred governance frameworks that balance innovation with worker protection; developing inclusive skills programmes with specific attention to women and youth populations vulnerable to disruption; supporting micro, small and medium enterprises to overcome barriers to AI adoption that often exclude smaller players from technological gains; and strengthening regional knowledge exchange and coordination on human resource development. These recommendations acknowledge that AI's impact will not be technologically determined but instead shaped by deliberate policy choices.

For Malaysian policymakers and workers, the ILO findings suggest both urgency and opportunity. Malaysia's economy, while less exposed than Singapore's and more exposed than Indonesia's, sits in a middle position requiring proactive management. The country must simultaneously invest in digital infrastructure and skills development to ensure workers can transition into AI-augmented roles rather than face displacement. Women's overrepresentation in high-exposure occupations demands gender-conscious reskilling initiatives that prevent AI from becoming another mechanism for gender-based employment discrimination. Companies, particularly in manufacturing and services sectors where Malaysia has historical strength, should begin integrating AI capability-building into their workforce development strategies now, before disruption accelerates.