When Kelvin Chiang presented his team’s agentic AI models to Singapore’s Monetary Authority, he knew he was demonstrating something unprecedented. What used to consume an entire workday for a private banker—compiling wealth reports, validating sources of funds, drafting compliance documents—now takes just 10 minutes. But before Bank of Singapore could deploy these tools across its wealth management division, Chiang’s data scientists had to walk regulators through every safeguard, every failsafe, and every human oversight mechanism designed to prevent the system from “hallucinating” false information.

The regulators didn’t push back. They embraced it.

That collaborative spirit between government and industry defines Singapore’s radically different approach to the AI transformation sweeping global banking. While financial institutions in the United States and Europe announce mass layoffs—Goldman Sachs warning of more job cuts as AI takes hold—Singapore is executing the world’s most ambitious banking workforce retraining program. DBS Bank, OCBC, and United Overseas Bank are retraining all 35,000 of their domestic employees over the next two years, a government-backed initiative that represents not just a skills upgrade, but a fundamental reimagining of what it means to work in financial services.

The Revolutionary Scale of Singapore’s AI Training Initiative

The numbers tell only part of the story. Singapore’s three banking giants are investing hundreds of millions in a training infrastructure that reaches from entry-level tellers to senior executives. But unlike generic technology upskilling programs that plague many organizations, this bootcamp targets specific, measurable competencies needed to work alongside autonomous AI systems.

Violet Chung, a senior partner at McKinsey & Company, identifies what makes this initiative unique: “The government is doing something about it because they realize that this capability and this change is actually infusing potentially a lot of fear.” That acknowledgment of worker anxiety—combined with proactive solutions rather than platitudes—sets Singapore apart from Western approaches that often prioritize shareholder returns over workforce stability.

The Monetary Authority of Singapore (MAS) isn’t just cheerleading from the sidelines. Deputy Chairman Chee Hong Tat, who also serves as Minister for National Development, has made workforce resilience a regulatory expectation. The message to banks is clear: deploy AI aggressively, but ensure your people evolve with the technology. Singapore’s National Jobs Council, working through the Institute of Banking and Finance, offers banks up to 90% salary support for mid-career staff reskilling—an unprecedented level of public investment in private sector workforce development.

Understanding Agentic AI: The Technology Driving the Transformation

To grasp why 35,000 bankers need retraining, you must first understand what agentic AI does differently than the chatbots and recommendation engines that preceded it.

Traditional AI systems respond to prompts. Ask a question, get an answer. Agentic AI, by contrast, pursues goals autonomously. According to research from Deloitte, these systems can plan multi-step workflows, coordinate actions across platforms, and adapt their strategies in real-time based on changing circumstances—all without constant human intervention.

Consider OCBC’s implementation. Kenneth Zhu, the 36-year-old executive director of data science and AI, oversees a lab where 400 AI models make six million decisions every single day. These aren’t simple calculations. The models flag suspicious transactions, score credit risk, filter false positives in anti-money laundering systems, and even draft preliminary reports that once consumed hours of compliance officers’ time.

At DBS Bank, an internal AI assistant now handles more than one million prompts monthly. The bank has deployed role-specific tools that reduce call handling time by up to 20%—not by replacing customer service staff, but by handling the tedious documentation and data retrieval that used to interrupt human conversations. Customer service officers now spend their time actually serving customers, while AI manages the administrative burden.

The source of wealth verification process at Bank of Singapore exemplifies agentic AI’s potential. Relationship managers previously spent up to 10 days manually reviewing hundreds of pages of client documents—financial statements, tax notices, property valuations, corporate filings—to write compliance reports. The new SOWA (Source of Wealth Assistant) system completes this same analysis in one hour, cross-referencing Bank of Singapore’s extensive database and OCBC’s parent company records to validate information plausibility.

Bloomberg Intelligence forecasts that DBS will generate up to S$1.6 billion ($1.2 billion) in additional pretax profit through AI-derived cost savings—roughly a 17% boost. These aren’t theoretical projections. DBS CEO Tan Su Shan reports the bank already achieved S$750 million in AI-driven economic value in 2024, with expectations exceeding S$1 billion in 2026.

Inside the Bootcamp: How 35,000 Bankers Are Actually Learning AI

The phrase “AI bootcamp” might conjure images of programmers teaching SQL queries. Singapore’s program looks nothing like that.

The curriculum divides into three tiers, each calibrated to job function and AI exposure level:

Tier 1: AI Literacy for Everyone (All 35,000 employees)

• Understanding what AI can and cannot do
• Recognizing AI-generated content and potential hallucinations
• Data privacy and security in AI contexts
• Ethical considerations when deploying automated decision-making
• Prompt engineering basics for interacting with AI assistants

Tier 2: AI Collaboration Skills (Frontline and Middle Management)

• Working with AI co-pilots for customer service
• Interpreting AI-generated insights and recommendations
• Overriding AI decisions when human judgment is required
• Monitoring AI system performance and reporting anomalies
• Translating customer needs into AI-friendly inputs

Tier 3: AI Development and Governance (Technical Teams and Senior Leaders)

• Model risk management frameworks
• Building and validating AI use cases
• Implementing responsible AI principles (fairness, explainability, accountability)
• Regulatory compliance for AI systems
• Strategic AI investment and ROI measurement

The Institute of Banking and Finance Singapore doesn’t just offer online modules. Through its Technology in Finance Immersion Programme, the organization partners with banks to create hands-on learning experiences. Participants work on actual banking challenges, developing practical skills rather than theoretical knowledge.

Dr. Jochen Wirtz, vice-dean of MBA programs at National University of Singapore, emphasizes the urgency: “Banks would be completely stupid now to load up on employees who they will then have to let go again in three or four years. You’re much better off freezing now, trying to retrain whatever you can.”

That philosophy explains why DBS has frozen hiring for AI-vulnerable positions while simultaneously training 13,000 existing employees—more than 10,000 of whom have already completed initial certification. Rather than the classic “hire-and-fire” cycle that characterizes American banking, Singapore pursues “freeze-and-train.”

The Human Reality: Fear, Adaptation, and Unexpected Opportunities

Not everyone welcomes their AI co-worker with open arms.

Bank tellers watching their branch traffic decline, back-office analysts seeing AI handle tasks they spent years mastering, relationship managers uncertain how to add value when machines draft perfect emails—the anxiety is real and justified. Singapore’s approach acknowledges these concerns rather than dismissing them.

Walter Theseira, associate professor of economics at Singapore University of Social Sciences, notes that banks are managing workforce transitions through “natural attrition rather than forced redundancies.” When employees retire, change roles internally, or move to other companies, banks increasingly choose not to backfill those positions. This gradual adjustment—combined with the creation of new AI-adjacent roles—softens the disruption.

The emerging job categories reveal how AI transforms rather than eliminates work:

• AI Quality Assurance Specialists: Testing AI outputs for accuracy, bias, and regulatory compliance
• Digital Relationship Managers: Handling complex wealth management with AI-generated insights
• Automation Process Designers: Identifying workflows suitable for AI augmentation
• Model Risk Officers: Ensuring AI systems operate within approved parameters
• Customer Experience Strategists: Designing human-AI interaction patterns

UOB has given all employees access to Microsoft Copilot while deploying more than 300 AI-powered tools across operations. OCBC reports that AI-assisted processes have freed up capacity equivalent to hiring 1,000 additional staff—capacity redirected toward higher-value customer interactions and strategic initiatives rather than eliminated.

One success story circulating in Singapore’s banking community involves a former transaction processor who completed the AI training program and now leads a team designing automated fraud detection workflows. Her deep understanding of payment patterns—knowledge that seemed obsolete when AI took over transaction processing—became invaluable when combined with technical AI literacy. She didn’t lose her job to automation; she gained leverage over it.

Singapore’s Regulatory Philosophy: Partnership Over Policing

What separates Singapore’s approach from virtually every other financial center is how its regulator, the Monetary Authority of Singapore, engages with AI deployment.

In November 2025, MAS released its consultation paper on Guidelines for AI Risk Management—a document that reflects months of collaboration with banks rather than top-down dictates imposed on them. The guidelines focus on proportionate, risk-based oversight rather than prescriptive rules that could stifle innovation.

MAS Deputy Managing Director Ho Hern Shin explained the philosophy: “The proposed Guidelines on AI Risk Management provide financial institutions with clear supervisory expectations to support them in leveraging AI in their operations. These proportionate, risk-based guidelines enable responsible innovation.”

The guidelines address five critical areas:

1. Governance and Oversight: Board and senior management responsibilities for AI risk culture
2. AI Risk Management Systems: Clear identification processes and accurate AI inventories
3. Risk Materiality Assessments: Evaluating AI impact based on complexity and reliance
4. Life Cycle Controls: Managing AI from development through deployment and monitoring
5. Capabilities and Capacity: Building organizational competency to work with AI safely

Rather than banning certain AI applications, MAS encourages banks to experiment while maintaining rigorous documentation of safeguards. When Kelvin Chiang presented his agentic AI tools, regulators wanted to understand the thinking process, the oversight mechanisms, and the escalation protocols—not to obstruct deployment, but to ensure responsible implementation.

This collaborative regulatory stance extends to funding. Through the IBF’s programs, Singapore effectively subsidizes workforce transformation, recognizing that individual banks cannot bear the full cost of societal-scale reskilling. PwC research shows organizations offering AI training report 42% higher employee engagement and 38% lower attrition in technical roles—benefits that justify public investment.

MAS Chairman Gan Kim Yong, who also serves as Deputy Prime Minister, framed the imperative at Singapore FinTech Festival: “It is important for us to understand that the job will change and it’s very hard to keep the same job relevant for a long period of time. As jobs evolve, we have to keep the people relevant.”

The ROI Case: Why Massive AI Investment Makes Business Sense

Singapore’s banks aren’t retraining 35,000 workers out of altruism. The business case for AI transformation is overwhelming—provided the workforce can leverage it.

DBS CEO Tan Su Shan described AI adoption as generating a “snowballing effect” of benefits. The bank’s 370 AI use cases, powered by more than 1,500 models, contributed S$750 million in economic value in 2024. She projects this will exceed S$1 billion in 2026, representing a measurable return on years of investment in both technology and people.

The efficiency gains manifest across every banking function:

Customer Service: AI handles routine inquiries, reducing average response time while allowing human agents to focus on complex problems requiring empathy and judgment. DBS’s upgraded Joy chatbot managed 120,000 unique conversations, cutting wait times and boosting satisfaction scores by 23%.

Risk Management: OCBC’s 400 AI models process six million daily decisions related to fraud detection, credit scoring, and compliance monitoring—work that would require thousands of additional staff and still produce inferior results due to human attention limitations.

Wealth Management: AI-powered portfolio analysis and market insights allow relationship managers at private banks to serve more clients at higher quality. What once required a team of analysts now happens in real-time, personalized to each client’s specific situation.

Operations: Back-office processing that once consumed entire departments now runs largely automated, with humans focused on exception handling and quality assurance rather than manual data entry.

According to KPMG research, organizations achieve an average 2.3x return on agentic AI investments within 13 months. Frontier firms leading AI adoption report returns of 2.84x, while laggards struggle at 0.84x—a performance gap that could determine competitive survival.

The transformation isn’t limited to cost savings. DBS now delivers 30 million hyper-personalized insights monthly to 3.5 million customers in Singapore alone, using AI to analyze transaction patterns, life events, and financial behaviors. These “nudges”—reminding customers of favorable exchange rates, suggesting timely financial products, flagging unusual spending—drive engagement and revenue while genuinely helping customers make better decisions.

Global Context: How Singapore’s Model Differs from Western Approaches

The contrast with American and European banking couldn’t be starker.

JPMorgan Chase CEO Jamie Dimon speaks enthusiastically about AI’s opportunities while the bank deploys hundreds of use cases. Yet JPMorgan analysts project global banks could eliminate up to 200,000 jobs within three to five years as AI scales. Goldman Sachs continues warning employees to expect cuts. The narrative centers on efficiency gains and shareholder value, with workforce impact treated as an unfortunate but necessary consequence.

European banks face different pressures. Strict labor protections make large-scale layoffs difficult, but they also complicate rapid workforce transformation. Banks attempt gradual transitions through attrition, but without Singapore’s comprehensive retraining infrastructure, displaced workers often struggle to find equivalent roles.

Singapore’s model succeeds through three unique factors:

1. Government-Industry Alignment The close relationship between MAS, the National Jobs Council, and major banks enables coordinated action impossible in more fragmented markets. When Singapore decides workforce resilience matters, resources flow accordingly.

2. Social Contract Expectations Singapore’s three major banks operate with implicit understanding that their banking licenses come with social responsibilities. Massive layoffs would trigger regulatory and reputational consequences, creating strong incentives for workforce investment.

3. Manageable Scale With 35,000 domestic banking employees across three major institutions, Singapore can execute comprehensive training that would be logistically impossible for American banks with hundreds of thousands of global staff.

Harvard Business Review analysis suggests Singapore’s approach, while difficult to replicate exactly, offers lessons for other nations: establish clear regulatory expectations around workforce transition, provide financial support for retraining, create industry-specific training partnerships, and measure success not just by AI deployment speed but by workforce adaptation rates.

The 2026-2028 Horizon: What Comes Next

As Singapore approaches the halfway point of its two-year retraining initiative, early results suggest the model works—but also highlight emerging challenges.

DBS has already reduced approximately 4,000 temporary and contract positions over three years, while UOB and OCBC report no AI-related layoffs of permanent staff. The banking sector is discovering that AI changes job composition more than job quantity, at least in the medium term.

The next wave of transformation will test whether current training adequately prepares employees. Gartner forecasts that by 2028, agentic AI will enable 15% of daily work decisions to be made autonomously—up from essentially zero in 2024. As AI agents gain more autonomy, the human role shifts from executor to orchestrator, requiring even higher-order skills.

MAS is already considering how to hold senior executives personally accountable for AI risk management, recognizing that autonomous systems create novel governance challenges. The proposed framework would mirror the Monetary Authority’s approach to conduct risk, where individuals bear clear responsibility for failures.

Singapore is also grappling with an unexpected challenge: Singlish, the local English creole, creates complications for AI natural language processing. Models trained on standard English struggle with Singapore’s unique linguistic patterns, requiring localized AI development—which in turn demands more sophisticated training for local AI specialists.

The broader implications extend beyond banking. If Singapore succeeds in demonstrating that massive AI deployment can coexist with workforce stability through strategic retraining, it provides a template for other industries and nations facing similar disruptions.

McKinsey estimates that AI could put $170 billion in global banking profits at risk for institutions that fail to adapt, while pioneers could gain a 4% advantage in return on tangible equity—a massive performance gap. Singapore’s banks, with their AI-literate workforce, position themselves firmly in the pioneer category.

Lessons for the Global Banking Industry

Singapore’s AI bootcamp experiment offers actionable insights for financial institutions worldwide:

Start with Culture, Not Technology: The most sophisticated AI fails if employees resist or misuse it. Comprehensive training that addresses fears and demonstrates value creates buy-in impossible to achieve through top-down mandates.

Partner with Government: Workforce transformation at this scale exceeds individual firms’ capacity. Public-private partnerships can distribute costs while ensuring industry-wide capability building.

Measure What Matters: Singapore tracks not just AI deployment metrics but workforce adaptation rates, employee satisfaction with AI tools, and the emergence of new hybrid roles. These human-centric measures predict long-term success better than pure technology KPIs.

Reimagine Rather Than Replace: The most successful AI implementations augment human capabilities rather than substituting for them. Relationship managers with AI insights outperform both pure humans and pure machines.

Invest in Adjacent Capabilities: AI literacy alone isn’t enough. Workers need complementary skills—critical thinking, emotional intelligence, creative problem-solving—that AI cannot replicate but can amplify.

Create New Career Paths: As traditional roles evolve, new opportunities in AI quality assurance, model risk management, and human-AI experience design create advancement paths for ambitious employees.

Accept Gradual Transition: Singapore’s two-year timeline, with flexibility for individual banks to move faster or slower based on their readiness, acknowledges that workforce transformation cannot be rushed without creating unnecessary disruption.

The Verdict: A Model Worth Watching

As the financial world watches Singapore’s unprecedented experiment, the stakes extend far beyond one nation’s banking sector. The question isn’t whether AI will transform banking—that transformation is already underway. The question is whether that transformation must inevitably create massive worker displacement, or whether strategic intervention can enable human adaptation at the pace of technological change.

Singapore bets on the latter possibility. By retraining all 35,000 domestic banking employees, by creating robust public-private partnerships, by developing comprehensive curricula that address both technical skills and existential anxieties, the city-state attempts to prove that the future of work doesn’t have to be a zero-sum battle between humans and machines.

Early returns suggest the model works. Banks report measurable productivity gains without mass layoffs. Employees initially resistant to AI training increasingly embrace it as they discover enhanced rather than diminished job prospects. Regulators fine-tune an approach that enables innovation while maintaining safety.

Yet challenges remain. Can retraining keep pace with accelerating AI capabilities? Will the job categories being created prove as numerous and lucrative as those being transformed? What happens to workers who cannot or will not adapt, despite comprehensive support?

These questions lack definitive answers. What Singapore demonstrates beyond doubt is that workforce transformation of this magnitude is possible—that major financial institutions can deploy cutting-edge AI aggressively while simultaneously investing in their people’s futures.

When historians eventually assess the AI revolution’s impact on work, Singapore’s banking sector bootcamp may be remembered as either a successful proof of concept that other nations and industries replicated, or as an admirable but ultimately isolated experiment that proved impossible to scale beyond a small, tightly integrated economy.

The next two years will tell us which.