Singapore’s Graduate Employment Challenge in the Age of Automation and Cautious Hiring
Executive Summary
A Goldman Sachs analysis published in February 2026 identified a structural deterioration in employment outcomes for college graduates in the United States. Industries that historically absorbed the largest share of degree holders — finance, information services, and professional and business services — have been shedding an average of 9,000 jobs per month since 2023, reversing a pre-pandemic trend in which those same sectors added 44,000 jobs monthly. The analysis attributes this primarily to a frozen labour market following the extraordinary 2021–2022 hiring surge, rather than to AI-driven displacement. Yet the warning has direct and nuanced relevance for Singapore, a city-state whose economic model is built almost entirely on producing and deploying knowledge workers.
This article examines how the global forces identified by Goldman Sachs intersect with Singapore’s unique structural conditions: its exceptionally high graduate density, its dependence on multinational headquarters employment, a rising tide of automation among PMET (Professional, Manager, Executive, Technician) roles, and a government-led upskilling infrastructure that is being tested at scale. The analysis draws on the latest data from Singapore’s Ministry of Manpower (MOM), SkillsFuture Singapore’s Graduate Employment Surveys, and the World Economic Forum’s Future of Jobs Report 2025.
6.8%
Public university graduates
unemployed within 6 months
(2024 cohort) 12.9%
Proportion of public university
grad labour force unemployed
(up from 5.6% in 2021)
74.8%
PEI graduates employed
within 6 months (2024)
vs. 83.2% in prior cohort 97%
Singapore employers reporting
significant AI exposure
(WEF Future of Jobs 2025, highest globally)
- Singapore’s Graduate Labour Market: A Portrait of Compression
1.1 A Historically Enviable Position — Now Under Pressure
Singapore has long achieved what many developed economies can only aspire to: near-full employment for university graduates, high starting salaries, and a skills ecosystem continuously upgraded through government intervention. The Joint Autonomous Universities Graduate Employment Survey (JAUGES) has historically reported employment rates within six months of graduation in the high eighties to low nineties in percentage terms. That architecture of graduate success is now showing structural cracks.
According to SkillsFuture Singapore’s Graduate Employment Survey data, the proportion of public university graduates in the labour force who were unemployed climbed from 5.6% in 2021 to 12.9% by 2024 — a 130% increase in three years. Full-time employment rates for fresh graduates have also declined across all institution types, even as median starting salaries continued to rise modestly. The 2024 cohort showed 6.8% of the 10,900 surveyed public university graduates were unemployed and actively seeking work, compared to 3.6% for the 2022 cohort.
It is important to contextualise this data carefully. As analysts have rightly noted, the extraordinarily low unemployment figures of 2021 and 2022 were anomalous — a product of pandemic-era workforce compression that temporarily removed a large share of non-resident workers from the Singapore labour market, inflating demand for locals. The pre-pandemic years (2016–2019) recorded unemployment-in-labour-force figures ranging from 9.3% to 11.1%, suggesting that the 2024 figure of 12.9% is elevated but not historically unprecedented. What is more concerning is the trajectory and the structural forces behind it.
“With AI enabling [employers] to do more with fewer people, employers have the luxury of hiring only the best — and often prefer experienced talent over fresh graduates.” — FYT Consultants, September 2025
1.2 The Salary-Employment Paradox
One of the most analytically interesting features of the current Singapore graduate labour market is the divergence between employment rates and starting salaries. Despite declining rates of full-time placement, median starting salaries for university graduates rose to SGD 4,500 per month in 2024, while polytechnic graduates earned a median of SGD 2,900. Law graduates commanded a median of SGD 7,000; Computer Science, Information Security, and Data Science graduates exceeded SGD 6,000.
This paradox — fewer jobs, but higher pay for those who secure them — reflects a bimodal market. Demand for top-tier technical talent in AI, data science, cloud architecture, and cybersecurity remains robust and intensely competitive, with AI/ML engineers reportedly earning between SGD 90,000 and SGD 170,000 annually. The market for general-purpose degree holders entering conventional PMET roles in operations, marketing, HR, or financial advisory, however, has weakened meaningfully. The Graduate Employment Survey data conflates these two very different experiences into aggregate figures, which obscures the depth of the structural challenge facing graduates in non-technical disciplines.
1.3 The Gap Between Vacancies and Access
Singapore’s overall vacancy-to-unemployed ratio remained at 1.35 as of mid-2025, suggesting that more jobs exist than unemployed persons. This headline figure has led some analysts to characterise the graduate unemployment question as primarily a skills-mismatch or expectations problem rather than a structural deficit. This reading deserves scrutiny. A vacancies-to-unemployed ratio measures availability, not suitability or accessibility. Many of the highest-vacancy roles as of June 2025 are in healthcare and social services — sectors that, while vital, are not the primary career destinations of NUS or NTU business or arts graduates, nor of the large cohort of fresh technology graduates now entering a more competitive pipeline.
The ManpowerGroup Employment Outlook Survey for Q1 2026 found Singapore’s Net Employment Outlook (NEO) stood at +15%, its lowest since Q1 2022. Nearly half of surveyed employers (46%) expected to maintain current headcount rather than expand it. In this environment, entry-level hiring competes directly against the option of not hiring at all — a calculus that consistently disfavours the candidate with no track record. - The Goldman Sachs Thesis and its Singapore Resonances
2.1 The Frozen Labour Market Hypothesis
The Goldman Sachs analysis, authored by economist Jessica Rindels, identifies what it terms a frozen labour market as the primary driver of deteriorating graduate employment outcomes. The mechanism is straightforward: the 2021–2022 period saw unprecedented front-loading of hiring in graduate-intensive sectors. Once those positions were filled, the demand pipeline dried up. Hiring freezes, quiet layoffs, and the absence of net new role creation have since locked recent graduates out of a market that remains structurally intact but is not growing.
This dynamic maps closely onto Singapore’s post-pandemic labour story. Singapore’s finance and insurance sector, information and communications, and professional services collectively employ a disproportionate share of PMET workers and university graduates. MOM data shows that resident employment growth in financial and insurance services, while still positive in mid-2025, has decelerated sharply. Outward-oriented industries — those most exposed to global economic uncertainty, including tech-adjacent consulting and professional services — showed weakening hiring sentiment through 2025. The sectors still adding jobs at scale are construction, healthcare, and social services: precisely those that recruit from a different skills profile.
2.2 The MNC Headquarters Problem
Singapore’s economic strategy has historically leveraged the Economic Development Board’s success in attracting multinational corporations to establish their regional headquarters in the city-state. This model created a deep pipeline of PMET roles for Singapore graduates and enabled the government to sustain a knowledge-intensive, high-wage economy despite the country’s lack of natural resources or domestic market scale.
There are now structural pressures on this model. Rising costs — Singapore’s salaries have increased faster than productivity growth in recent years — have made the city-state less competitive as an MNC hub compared to alternatives in Southeast Asia. Simultaneously, MNCs globally are responding to AI-enabled productivity gains by doing more with fewer people: hiring is shifting from capacity-building to quality-selective replacement. When Amazon cuts 14,000 white-collar roles or Microsoft reduces headcount citing AI productivity gains, the downstream effect for Singapore is a reduction in the PMET hiring that occurs within their Singapore operations.
The SkillsFuture analysis of Singapore’s banking sector confirms this shift. Between 2019 and 2024, major domestic systemically important banks (DBS, OCBC, UOB, and their peers) and smaller firms both reduced hiring in IT Support and Web and Mobile Applications Developer roles. The shift was away from active digital product development toward the maintenance and optimisation of mature infrastructure — a signal that one wave of digital transformation has been absorbed and the next, AI-driven wave has not yet produced equivalent net hiring.
2.3 Is AI the Real Driver in Singapore?
The Goldman Sachs analysis is careful to argue that AI has not been the primary driver of graduate unemployment in the United States, at least not yet. The same claim is broadly defensible for Singapore, but requires a more granular disaggregation than the headline numbers offer.
The World Economic Forum’s Future of Jobs Report 2025 found that 97% of Singapore employers reported significant AI exposure — the highest figure globally and notably higher than the US or European averages. The WEF estimated that nearly three in ten jobs in Singapore would face structural labour-force churn over the next five years, with a 36% skill disruption rate. Against this backdrop, the argument that AI is not yet materially affecting graduate hiring in Singapore may be technically accurate in terms of announced layoffs attributable to AI, but structurally misleading. The more significant effect is on hiring intent: AI enables firms to accomplish with existing staff what previously required fresh headcount. The job is not eliminated; it is simply not opened.
NUS Business School’s Professor Jochen Wirtz has noted that even low-level PMET roles involving call centre operations and back-office paperwork are now subject to end-to-end automation. This creates a compression effect at the entry level, where first-job roles historically served as the on-ramp through which graduates established the professional networks and credentials that enable mid-career transitions. Eliminating or contracting this on-ramp has compounding effects that will take years to fully manifest in labour market data. - Sector-by-Sector Analysis
3.1 Finance and Banking
Singapore’s banking and financial services sector employs approximately 200,000 professionals, representing about 5% of the national labour force. It remains the strongest-performing sector in terms of hiring sentiment: ManpowerGroup’s Q1 2026 survey identified Finance and Insurance as the sector with the highest NEO at +33%. However, this figure needs careful interpretation. Sector-level growth is concentrated in regulatory compliance, risk management, sustainability-linked finance, and AI implementation roles — not the broad-based analyst and relationship manager hiring that once absorbed large cohorts of business and economics graduates.
The SkillsFuture data shows that both major banks and smaller financial institutions have been reducing their share of Financial/Investment Adviser roles, reflecting both digital disruption of retail banking and a shift in consumer financial behaviour. Demand for Risk Management Managers, by contrast, has remained resilient — a reflection of the increasing complexity of the regulatory and ESG environment. The net effect for new graduates is that the finance sector continues to hire, but for a narrower band of specialised profiles.
3.2 Information and Communications Technology
Singapore’s ICT sector remains the epicentre of both opportunity and displacement anxiety. On one hand, AI/ML engineering, data science, cloud architecture, and cybersecurity roles command premium salaries and face genuine talent shortages. On the other, the sector has been contracting at the entry level: outward-oriented ICT firms, heavily exposed to global technology spending cycles, have been hiring less even as they remain the primary employers of skills-intensive graduates.
The AI Singapore Apprenticeship Programme (AIAP), a nine-month government-funded initiative providing hands-on experience with real-world AI projects, is free but highly selective. SkillsFuture credits can be deployed toward NUS SCALE and NTU PACE certification programmes in data science and AI. These institutional pathways are valuable but represent a bottleneck: the programmes are competitive and time-intensive, and they target a graduate profile that already possesses foundational technical knowledge. Graduates from non-technical disciplines seeking to pivot into ICT face a more arduous transition, despite the availability of retraining infrastructure.
3.3 Professional and Business Services
Consulting, legal services, accounting, and HR represent a second major graduate employment pipeline that is undergoing quiet but significant restructuring. Generative AI tools can now perform preliminary legal research, produce first-draft contracts, conduct financial modelling, screen candidates, and produce market research briefs. These are precisely the tasks historically performed by junior associates, analysts, and fresh graduate trainees — roles that also served as the primary mechanism through which firms assessed graduate talent before committing to longer-term employment.
The contraction of these entry-level roles does not show up in large-scale announced layoffs. Instead, it manifests as a reduction in graduate intake, a shift from permanent positions to fixed-term contracts, and a preference for experienced hires who can operate independently without the productivity cost of training. SkillsFuture’s May 2025 survey noted that more employers are favouring fixed-term contracts over full-time permanent roles for entry-level candidates — a structural shift that increases graduate labour market fragility without registering clearly in unemployment statistics.
3.4 Healthcare and Social Services
In contrast to the contracting graduate employment pipeline in white-collar services, healthcare and social services represents one of the fastest-growing areas of vacancy creation in Singapore. MOM data through Q2 2025 confirmed that health and social services had the largest number of open vacancies, driven by Singapore’s rapidly ageing population demographics. The peak working-age population was reached in 2023; the dependency ratio will only increase through the 2030s.
This creates a structural opportunity that is only partially being captured. The sector can absorb graduates from nursing, social work, and allied health programmes, but the salary bands — while improving under Singapore’s Progressive Wage Model — remain lower than equivalent-level roles in finance or technology. This creates a preferences mismatch: graduates from higher-ranked universities with higher salary expectations are less likely to enter healthcare roles, even when vacancies exist. The result is a paradox in which genuine labour shortages coexist with graduate unemployment in the same economy. - The Structural Vulnerabilities of Singapore’s Graduate Economy
4.1 Concentration Risk and the MNC Dependency
Singapore’s economic model involves a high degree of concentration risk with respect to graduate employment. The PMET category, which includes most university graduates, is disproportionately employed in MNC-affiliated roles in finance, professional services, and technology. This concentration means that when global MNCs enter a period of white-collar hiring restraint — as they have done since 2023 — the effect on Singapore’s graduate employment market is amplified relative to more diversified economies.
Unlike the United States, Singapore does not have a large domestic market that can generate independent demand for knowledge-worker services. Its domestic small and medium enterprises (SMEs), which account for the majority of business establishments, are primarily in trade, retail, food and beverage, and construction — sectors that are growing but that absorb a different skills profile. The concentration of graduate employment opportunity within the MNC and financial services ecosystem creates a structural fragility that Goldman Sachs’s frozen labour market thesis exposes with particular sharpness.
4.2 The PMET Squeeze and Wage Expectations
Singapore’s PMET employment challenge is qualitatively different from the blue-collar displacement story that dominates international coverage of automation. The threat is not primarily to manual roles but to the analytical, coordinative, and communicative functions that have historically constituted white-collar work. AI can screen resumes, forecast sales, optimise advertising campaigns, draft compliance documents, and produce data summaries. These functions are not the entire content of a PMET role, but they are precisely the functions that junior-level staff are hired to perform in their first years of employment.
The resulting pressure creates what analysts have called the PMET squeeze: a compression of opportunity at the entry level, combined with rising productivity expectations for those who do enter. Singapore’s business community has raised anecdotal concerns about the work ethic and resilience expectations of younger workers, but this cultural framing obscures a structural reality. The economic circumstances that enabled previous generations of graduates to accept low-paid junior roles — in the expectation of wage growth as skills accumulated — are changing. When AI handles the routine tasks and firms hire selectively for higher-level functions, the entry-level training ground that once existed no longer operates on the same terms.
4.3 The Small State Vulnerability in AI Disruption
The Civil Service College of Singapore has noted that small states like Singapore face particular vulnerability in the current wave of technological change. Global powers are exerting influence on technology supply chains, including components critical to AI innovation such as advanced chip manufacturing. Singapore’s dependence on imported AI infrastructure and on technology developed by US and Chinese hyperscalers creates a structural constraint on its ability to shape the tools that will reshape its own labour market.
Singapore currently lacks the scale to develop large foundation AI models domestically. Its AI Singapore initiative focuses on deployment and application, not on creating the underlying models. This is economically rational for a small state but means that the pace and direction of AI capability development — which will determine how rapidly tasks currently performed by PMET workers can be automated — is determined externally. This is a strategic vulnerability that no amount of domestic reskilling policy can fully address. - Singapore’s Policy Response: Strengths, Gaps, and Limits
5.1 The SkillsFuture Architecture
Singapore’s SkillsFuture initiative, launched in 2015, represents one of the most comprehensive government-funded lifelong learning ecosystems globally. The initiative provides subsidised access to thousands of courses, career transition support, Skills Ambassadors for personalised skills consultation, and the Jobs-Skills Portal for mapping skills demand to career pathways. Budget 2025 introduced additional workforce support measures targeted specifically at fresh graduates and mid-career workers navigating economic uncertainty.
The SkillsFuture Skills Demand for the Future Economy (SDFE) 2025 report identifies three priority economic pillars — the Care Economy, the Digital Economy, and the Green Economy — and maps the skill trajectories most likely to generate sustainable employment. The analytical infrastructure is world-class. The challenge is that knowledge of what skills are in demand is necessary but not sufficient. The structural problem is that the volume of entry-level PMET opportunity is contracting independently of skills provision.
5.2 The Training-to-Placement Gap
Singapore’s upskilling infrastructure is well-designed for a world in which graduates and mid-career workers need skills updates to transition from declining to growing roles within a broadly intact labour market. It is less well-designed for the possibility that the aggregate number of roles appropriate for degree-level knowledge workers may structurally decline, or that the primary mechanism of displacement is not layoffs (which are visible and respond to retraining) but hiring suppression (which is invisible and does not).
When firms use AI to avoid creating new roles, there is no retraining pathway — because the role that would have been created was never announced. The graduate is not displaced from a job; they are simply never given a foot in the door. SkillsFuture’s emphasis on making workers more productive and more upskillable addresses the displacement risk well. It does not directly address the creation risk — the question of whether the hiring of new entrants will recover to pre-2023 levels as AI productivity compounds.
5.3 The Tripartite Model Under Strain
Singapore’s labour market has been historically managed through a tripartite framework involving the government, employers represented by the Singapore Business Federation, and unions under the National Trades Union Congress. This model has been one of Singapore’s most durable competitive advantages, enabling coordinated wage adjustments, managed restructuring, and fast-moving policy responses to external economic shocks.
The tripartite model faces new strains in the AI era. The primary mechanism of AI-driven labour market disruption — selective hiring contraction and role non-creation — is harder to address through the traditional instruments of wage negotiation and workforce restructuring support. The Progressive Wage Model (PWM), which mandates minimum wages and career advancement pathways for certain lower-wage occupations, has been effective in its target sectors. It does not extend to the white-collar PMET categories where graduate employment pressure is most acute. The challenge for the tripartite framework is to develop analogous tools for a graduate employment market that is experiencing pressure not through explicit restructuring but through gradual demand suppression.
5.4 Singapore’s Smart Nation 2.0 Positioning
Singapore has deliberately framed AI not as a corporate efficiency tool but as a societal asset. Smart Nation 2.0 treats AI as a shared national resource meant to elevate citizens, and surveys consistently show that Singaporean workers have lower AI anxiety than their counterparts in markets like the UAE — a consequence of the embedded trust between government and citizens that Singapore’s governance model has cultivated.
This cultural positioning is a genuine strength. A population that approaches AI as a productivity partner rather than an existential threat is better positioned to develop the human-AI collaborative competencies that employers in growing sectors are seeking. However, positive framing does not resolve the structural problem: if the jobs available at the entry level for new graduates are fewer in number and narrower in scope, no amount of AI fluency training changes the underlying supply-demand calculus. - The Long-Term Picture: Three Scenarios
Scenario A: Cyclical Recovery
The most optimistic scenario holds that the deterioration in graduate employment outcomes since 2022 is primarily cyclical rather than structural — a consequence of the frozen labour market dynamic identified by Goldman Sachs, compounded by global economic uncertainty and geopolitical volatility. In this scenario, as the post-pandemic hiring overhang is absorbed, MNCs resume normal hiring cadences, and the graduate employment rate recovers toward historical norms. Singapore’s shrinking working-age population — the peak was reached in 2023 — provides an additional tailwind, gradually tightening the labour market as natural attrition outpaces new entrant volume. The historical resilience of degree holders in occupational transitions supports this reading.
Scenario B: Structural Compression
A more pessimistic scenario holds that AI-enabled productivity gains will permanently suppress the aggregate number of entry-level PMET roles, creating a permanently smaller on-ramp for new graduates. In this scenario, the salary premium for university graduates does not disappear — the top quartile of graduates continue to secure well-remunerated positions — but the middle of the distribution faces chronic underemployment relative to qualification levels and salary expectations. The private return to a bachelor’s degree narrows, pressure grows on universities to demonstrate employment relevance, and the social contract built around educational credentialism frays.
This scenario is consistent with the narrowing of Singapore’s historical PMET wage premium for young workers, the decline in full-time permanent graduate placements, and the shift toward fixed-term contracts. It implies that SkillsFuture-style interventions, while valuable, address only part of the challenge.
Scenario C: Structural Transformation
A third scenario — the most transformative — holds that AI creates a net increase in knowledge-intensive work over the medium term, but that the composition of that work shifts substantially. The WEF Future of Jobs Report 2025 projects that AI will create 170 million new roles globally while eliminating 92 million, for a net positive of 78 million by 2030. Singapore, with its AI-ready infrastructure and governmental commitment to digital transition, is well-positioned to capture a disproportionate share of the roles created.
In this scenario, the current period of graduate employment pressure is the painful transitional phase between the old PMET economy and the new AI-augmented economy. The risk is that the transition imposes concentrated costs on a generation of graduates who enter the labour market precisely during the structural adjustment — a cohort effect that no amount of individual upskilling can entirely neutralise. - Implications and Recommendations
For Policymakers
Measure the right thing: Current graduate employment metrics focus on employment within six months of graduation. A more informative metric would track trajectory — quality of first employment, wage growth after five years, and occupational mobility — to distinguish genuine structural deterioration from cyclical delays.
Address entry-level role creation directly: Consider designing incentives for firms that create net new entry-level PMET roles, rather than focusing exclusively on retraining supply. The demand side of the graduate employment equation needs direct policy attention.
Extend the tripartite model upward: The Progressive Wage Model’s success in lower-wage sectors suggests that coordinated minimum standards for graduate employment — including pathways from fixed-term to permanent status — could be developed for white-collar entry-level roles.
Invest in domestic AI capability: Singapore’s dependency on externally-developed AI infrastructure is a long-term strategic risk. Increasing domestic AI research capacity, even at modest scale, reduces the degree to which Singapore’s labour market is determined by decisions made in San Francisco and Beijing.
For Graduates and Educators
Develop AI complementarity, not just AI literacy: The key distinction is between roles where AI substitutes for human judgement and roles where AI amplifies it. Graduates who can work with AI tools at a sophisticated level — validating outputs, applying contextual human judgement, managing AI-augmented workflows — will consistently outcompete those who cannot.
Recalibrate sector expectations: The historical prestige hierarchy of graduate employment — finance and consulting at the top, public sector and healthcare below — no longer maps cleanly onto labour market opportunity. Healthcare, green finance, and AI governance are growing in both vacancy volume and long-term career potential.
Treat fixed-term contracts as productive entry points: The shift from permanent to contract-based entry-level hiring reflects employer risk aversion under uncertainty, not a permanent reduction in career opportunity. Graduates who demonstrate high performance on contract engagements have historically converted to permanent status.
For Employers
Recognise the pipeline externality: Individual firms that suppress entry-level hiring collectively degrade the talent pipeline on which the broader economy depends. When no firm hires junior analysts, no firm benefits from mid-career analysts with foundational experience. The frozen labour market dynamic is self-reinforcing and economically destructive in aggregate even when rational at the individual firm level.
Invest in structured graduate development: Singapore’s tripartite framework, including Workforce Singapore’s traineeship schemes and SkillsFuture Enterprise Credits, provides direct financial support for firms that invest in graduate training. These instruments are underutilised relative to their potential.
Conclusion
The Goldman Sachs warning about shifting labour market dynamics for college graduates arrives in Singapore at a moment of genuine structural uncertainty. Singapore’s graduate employment metrics have deteriorated meaningfully since 2022, and the forces driving that deterioration — a frozen post-pandemic hiring cycle, AI-enabled productivity gains that suppress role creation, and geopolitical uncertainty that makes MNCs cautious about headcount expansion in expensive locations — are not cyclical noise. They reflect real changes in the economics of graduate employment.
Singapore is better positioned than most economies to navigate this transition. Its SkillsFuture infrastructure is world-class, its tripartite labour relations model provides coordination mechanisms unavailable elsewhere, and its population’s high AI adoption and relatively low AI anxiety create a cultural foundation for productive human-AI collaboration. The Smart Nation 2.0 vision provides a coherent national narrative for the transition.
But these strengths are not unconditional. They address the supply side of the graduate employment challenge — producing adaptable, well-trained workers — more robustly than the demand side — ensuring that sufficient entry-level PMET roles exist for those workers to enter. If the frozen labour market dynamic persists, or if AI-enabled productivity gains compress the aggregate demand for entry-level knowledge work faster than new role categories emerge, Singapore’s educational investment will produce graduates who are well-trained for a labour market that no longer exists in the form they were trained for.
The historical record suggests that Singapore has navigated structural labour market transformations before — from entrepot trade to manufacturing to finance to the knowledge economy — and done so with less economic disruption than peers. There is reasonable ground for optimism. But the window for policy intervention is narrowing, and the consequences of misdiagnosis — of treating a structural challenge as a cyclical correction — will be felt most acutely by the cohort of graduates entering the labour market in the next three to five years.
Sources and Data
Goldman Sachs Economics Research, Jessica Rindels (February 2026); Ministry of Manpower Singapore, Labour Market Advance Releases Q1–Q4 2025; SkillsFuture Singapore, Private Education Institution Graduate Employment Survey 2023/2024; Joint Autonomous Universities Graduate Employment Survey (JAUGES) 2024; World Economic Forum, Future of Jobs Report 2025; ManpowerGroup Employment Outlook Survey Q1 2026; Reeracoen Singapore Salary Guide 2026; FYT Consultants Labour Market Analysis (September 2025); Civil Service College Singapore, AI & Technology Report; Bureau of Labor Statistics (US), December 2025.