Saudi Arabia vs Yemen: AI Job Risk Across a War Divide

Two countries, one border, a 54x economic divide

Saudi Arabia and Yemen share the Arabian Peninsula's longest land border - approximately 1,800 kilometres. They share cultural and historical ties stretching back millennia. In 2026, they share almost nothing economically. Saudi Arabia's GDP per capita of $34,537 (World Bank 2025) places it in upper-middle to high-income territory. Yemen's GDP per capita of $634 (World Bank 2025) makes it one of the poorest countries in the world, and the poorest in the Arab world. The 54x gap between them is not a development gap - it is the outcome of a decade of war.

Saudi Arabia tracks 14,192,000 workers via ILO ILOSTAT (CC BY 4.0), 2025 data year. Yemen's most recent ILO ILOSTAT data is from 2010 (CC BY 4.0) - a 2010 Labour Force Survey conducted five years before the civil conflict began. The 4,092,000 workers counted in that survey represent pre-war Yemen. The current Yemeni workforce is smaller, more informally employed, more heavily agricultural and subsistence-oriented, and vastly less formally tracked than the 2010 figure suggests. Any analysis of Yemen's workforce must carry this disclaimer prominently.

Side-by-side: occupation groups in both countries

Saudi Arabia: ILO ILOSTAT (CC BY 4.0), 2025 data year. Yemen: ILO ILOSTAT (CC BY 4.0), 2010 data year - pre-civil war. Neither country has OECD wage data available at any ISCO-1 level.

Saudi Arabia ILO ILOSTAT 2025 does not separately enumerate skilled agricultural and elementary occupations at ISCO-1 level for this data year. Yemen figures are from the 2010 Labour Force Survey - pre-civil war data that does not reflect current conditions. No OECD wage data is available for either country.

Saudi Arabia's Vision 2030 and the AI exposure question

Saudi Arabia's average AI exposure of 4.82/10 places it above many comparable Middle Eastern economies. The professional group at 3,504,800 workers (24.7% of the workforce, score 6.5/10) and technicians at 2,347,100 (16.5%, score 5.5/10) together account for 5,851,900 workers - 41.2% of the tracked workforce - all scoring 5.5 or above.

An important context for the Saudi workforce figure: approximately 38% of Saudi Arabia's workforce consists of expatriate workers (Migrant Forum in Asia, citing Saudi General Authority for Statistics data). ILO ILOSTAT tracks all employed residents, so the 14,192,000 figure includes Saudi nationals and foreign nationals working in the Kingdom. The occupation breakdown reflects this mixed composition - a significant share of professionals, technicians, and service workers in Saudi Arabia are foreign nationals from South Asia, Southeast Asia, and Arab countries working under kafala contracts. AI displacement dynamics for these workers differ from Saudi nationals, as policy decisions around Saudisation (Nitaqat programme quotas) interact with automation decisions in complex ways.

Vision 2030 - Saudi Arabia's national transformation plan launched in 2016 - explicitly prioritises AI and technology adoption as drivers of economic diversification away from oil dependence. The Saudi Data and AI Authority (SDAIA) was established in 2019 and has pursued large-scale AI deployment across government services, healthcare, and financial services. This active investment in AI infrastructure pushes Saudi Arabia's risk velocity to 1.0/10 in the dataset - low in absolute terms reflecting the workforce's still-heavy reliance on manual and services labour, but reflecting genuine government-backed AI deployment that most comparable economies at this income level have not undertaken.

Yemen: when AI risk velocity is zero

Yemen's risk velocity score of 0.0/10 is the lowest possible score. It is assigned to countries where the conditions required for AI adoption - stable electricity supply, functioning internet infrastructure, employer capital, a working formal economy, and physical security for businesses to operate - are not present. Yemen has none of these in 2026.

Yemen's civil war, which escalated in 2015 with Saudi-led coalition military intervention following Houthi forces seizing Sanaa, has caused what the United Nations has repeatedly described as one of the world's worst humanitarian crises. As of 2026, Yemen remains without a unified government or functioning national electricity grid. Internet penetration, while growing in Houthi-controlled areas through satellite connectivity, is not at a level that supports enterprise software deployment. The formal banking system is fragmented between competing authorities. Most of the economy that functions at all does so informally - subsistence agriculture, local trade, and remittances from Yemeni workers abroad.

In this context, the AI exposure scores derived from the 2010 Labour Force Survey data are better understood as a historical baseline than a current risk assessment. Yemen's pre-war skilled agricultural workforce of 870,800 workers (21.3% of the 2010 total, score 3.0/10) has likely grown as a share of the surviving economy as urban formal employment has collapsed. Yemen's armed forces occupations at 352,200 workers in 2010 (score 2.5/10) reflect a pre-war national army - the current situation involves multiple armed factions and a militarised economy that ILO data cannot capture.

Economy context: Saudi Arabia vs Yemen

The table below uses World Bank Open Data (CC BY 4.0, 2025) and UNDP Human Development Report 2025 (HDR 2025, 2023 data year, licence CC BY 3.0 IGO).

Yemen's World Bank-reported unemployment rate of 17.32% (2025) should be read carefully. The World Bank derives this from modelled estimates rather than current Labour Force Survey data - Yemen has not conducted a reliable national employment survey since before the war. The real unemployment and underemployment rate in Yemen is believed to be substantially higher, with the International Labour Organization estimating pre-war formal sector employment has largely collapsed in Houthi-controlled governorates and severely contracted elsewhere. Yemen's 0.470 HDI (rank 184, UNDP HDR 2025) - the second-lowest in the world - reflects the aggregate impact of conflict on health, education, and living standards.

What this means for Saudi workers

For Saudi Arabia's professionals (3,504,800 workers, score 6.5/10), the AI disruption question is real and increasingly near-term. Vision 2030 has driven active deployment of AI tools in government services (the Saudi Digital Government Authority has announced multiple AI automation programmes), healthcare (King Faisal Specialist Hospital and other major institutions are deploying AI diagnostic tools), and financial services (Saudi banks and fintech firms are among the most active AI adopters in the GCC). Saudi professional workers in administrative functions, financial analysis, and healthcare administration face the same augmentation and displacement dynamics as professionals in more developed economies.

The Saudisation dimension adds complexity. Saudi Arabia's Nitaqat programme requires private sector firms above certain sizes to employ minimum percentages of Saudi nationals. These quotas interact with AI adoption in non-obvious ways: if AI tools enable one Saudi employee to do the work of three, employers may meet Saudisation quotas with fewer total employees - reducing expatriate headcount but potentially also reducing the total number of Saudis employed at middle levels as senior Saudis use AI to manage larger portfolios. The net effect on Saudi national employment from AI-plus-Saudisation is not straightforwardly positive.

For Saudi Arabia's clerical workers (848,000 at 8.5/10), near-term risk is real. Saudi government agencies and large private employers in banking, insurance, and telecoms are among the most active in the region in deploying document processing, correspondence, and administrative workflow automation. These 848,000 workers are the group facing the most immediate exposure in the Saudi workforce.

What this means for Yemeni workers

For Yemeni workers, the AI exposure score is the least relevant number in this comparison. The real constraints facing Yemen's workforce in 2026 are conflict, displacement, malnutrition, collapsed healthcare, destroyed infrastructure, and a fragmented economy. The United Nations Office for the Coordination of Humanitarian Affairs (OCHA) estimates that approximately 21.6 million Yemenis need humanitarian assistance (OCHA Yemen Situation Report 2025). In that context, AI displacement is not a near-term or medium-term concern - it is a question for a post-conflict reconstruction period whose timeline is deeply uncertain.

Yemen's pre-war skilled agricultural workers (870,800 at 21.3% of the 2010 workforce, score 3.0/10) and the broader agricultural and subsistence economy represent the labour market most likely to survive conflict conditions, precisely because subsistence farming does not depend on formal employer capital, internet connectivity, or functioning financial systems. These workers face the lowest AI exposure in the pre-war dataset - and they face the most human-scale immediate challenges that have nothing to do with AI.