A new industry report finds that 86% of Singapore fintechs have deployed AI, yet most value its ability to reduce costs and manage risk. Data governance, talent shortages, and limited adoption in high-risk functions remain barriers to wider deployment.
Singapore's fintech sector is adopting artificial intelligence (AI) faster than many financial institutions, but turning that investment into new revenue remains a major challenge.
A new report, The AI Inflection Point: Rewiring Fintech, finds that 86% of Singapore fintechs have deployed AI. However, only 4% identify new revenue generation as the technology's biggest business benefit.
The report was launched on 18 November 2026 by the Singapore FinTech Association (SFA), global financial technology company Airwallex and fintech research firm Kapronasia at the Singapore FinTech Festival 2026.
Based on a survey of 100 Singapore fintech leaders and interviews with 10 senior industry practitioners, the research examines how deeply companies have integrated AI, where they see value, and what prevents wider adoption.
The findings reveal a clear gap between adoption and business impact. Most fintechs are using AI to reduce operating costs, manage risk, and improve employee productivity. However, fewer have translated these capabilities into revenue-generating activities or more advanced autonomous operations.
Singapore Fintechs Lead AI Adoption, but Integration Remains Uneven
Singapore fintechs reported an AI deployment rate of 86%, compared with 65% among Singapore financial institutions and 62% among global fintechs.
These comparisons draw on separate research sources cited in the report. Nevertheless, they place Singapore's fintech ecosystem ahead of the reported benchmarks.
However, high adoption does not necessarily mean deep integration.
The research found that 99% of respondents had deployed AI in technology and engineering. Operations and back-office functions followed at 98%.
Yet adoption drops sharply in higher-risk business areas.
Around 59% of respondents reported no AI adoption in financing, lending, and underwriting. Furthermore, only 29% reported integrated AI use in that function.
This difference reflects the risks involved in financial decisions. AI can help employees process information and analyze applications. However, lending decisions can directly affect customers and financial outcomes.
Consequently, firms appear more cautious about giving AI greater responsibility in these areas.
The report also found that 58% of respondents said fewer than half their employees used AI in daily work. This suggests that access to AI tools has not yet translated into broad workforce adoption.
Only 4% Identify New Revenue as AI's Biggest Benefit
The most striking finding concerns how fintechs measure AI's business value.
Operating cost reduction was the leading benefit, cited by 41% of respondents. Another 28% identified risk mitigation and compliance, while 19% pointed to employee efficiency and faster output.
Customer retention and experience accounted for 8%.
New revenue generation ranked last at just 4%.
<text weight="semibold" color="default">Where Singapore fintechs see the greatest value from AI</text>
|
Business benefit |
Share of respondents |
|---|---|
|
Operating cost reduction |
41% |
|
Risk mitigation and compliance |
28% |
|
Employee efficiency and output speed |
19% |
|
Customer retention and experience |
8% |
|
Net-new revenue creation |
4% |
The figures do not mean AI generates no additional revenue. Instead, they show that very few respondents consider new revenue its biggest current benefit.
One reason may be the difficulty of linking revenue growth directly to AI. Sales results depend on several factors, including pricing, customer demand, distribution, and product quality.
By contrast, firms can often measure cost savings and productivity improvements more directly.
The report therefore recommends measuring AI against outcomes owned by individual business functions. For example, sales teams can track deals closed, while operations teams can measure service-level agreement throughput.
Airwallex follows this approach by measuring AI through business outcomes rather than adoption numbers alone.
AI Investment Is Rising Despite a Return Gap
Singapore fintechs remain committed to expanding their AI capabilities.
The report found that 94% of respondents expect their AI budgets to increase in 2027. This signals continued investment despite challenges in measuring returns.
However, expected and reported returns do not fully match.
Around 61% of respondents expected returns of between 26% and 50% from AI investments. Only 42% reported achieving returns within that range.
Meanwhile, 45% reported returns between 1% and 25%. Only 21% had initially expected returns in that lower range.
<text weight="semibold" color="default">Expected versus reported AI investment returns</text>
|
Return range |
Expected |
Reported actual |
|---|---|---|
|
1%–25% |
21% |
45% |
|
26%–50% |
61% |
42% |
|
51%–100% |
9% |
1% |
The results highlight a challenge for fintech management teams. Companies are increasing spending, but many have not achieved the returns they initially expected.
The report also cautions against relying only on traditional return-on-investment measures. AI can improve product development speed, customer experience, and operational quality. These benefits may not appear immediately in revenue or cost figures.
The next phase will require firms to connect AI deployment with clear business objectives.
Data Governance Is the Biggest Barrier to Scaling AI
Technical readiness is not the only factor determining how quickly fintechs can expand AI.
The report identifies internal data governance as the biggest barrier, cited by 63% of respondents.
Cost and talent shortages each affected 59% of firms. Legacy technology integration followed at 55%, while trust in algorithmic decisions was cited by 41%.
This highlights a shift in the industry's main challenge. Companies have access to AI technology, but they still need reliable data, skilled teams, and effective controls.
Singapore fintechs show strong awareness of existing governance initiatives.
The research found that 81% were engaging with the Infocomm Media Development Authority's Model AI Governance Framework. Another 86% were engaging with the Monetary Authority of Singapore's Project MindForge.
However, implementation remains limited.
Only 9% reported putting controls in place aligned with the IMDA framework. The same proportion had established operating procedures and safeguards under Project MindForge.
This gap between awareness and implementation could become more important as AI systems gain access to sensitive financial data and operational processes.
Companies will need clear rules covering data access, model monitoring, accountability, and human intervention.
Agentic AI Remains at an Early Stage
The report also examines the industry's transition towards agentic AI.
Unlike conventional assistants, agentic systems can perform multi-step tasks, use tools, and take actions with less direct human involvement.
Despite growing interest, deployment remains limited.
Around 42% of surveyed fintechs said they were ready to implement agentic AI. However, only 11% had deployed it in live environments.
The gap reflects concerns about trust, accountability, and operational risk.
For example, an AI assistant can recommend a payment or identify a potential issue. An autonomous agent could go further by initiating transactions or managing financial workflows.
That additional authority requires stronger safeguards.
The report found that 74% of respondents considered human checkpoints necessary for agentic AI. Other important controls included activity logging at 53%, model monitoring and audits at 52%, and identity and authentication at 52%.
Approval and transaction limits were cited by 49%.
The findings suggest that financial institutions are more comfortable with AI supporting employees than independently executing sensitive financial decisions.
Endowus Shows How AI Can Improve Customer Support
The report also highlights practical examples of AI delivering operational benefits.
Singapore-based wealth management platform Endowus uses an AI-powered WhatsApp and in-app support bot to handle approximately 30% of customer queries.
This allows advisers and support teams to spend more time on complex requests and customer relationships.
The report notes that faster resolution times have contributed to improved client satisfaction. It does not attribute the improvement simply to AI outperforming human responses.
This example illustrates a practical approach to AI adoption. Firms can automate routine work while keeping human expertise available for situations that require judgment and personal attention.
The report also discusses Cleanverse's use of autonomous agents for prospect discovery and engagement. Meanwhile, Airwallex describes agentic business account capabilities designed to support financial operations within business-defined permissions and approvals.
These examples show how the technology could move beyond assistance towards more integrated workflows.
The Next AI Challenge Is Business Transformation
The report recommends three priorities for fintechs seeking greater value from AI.
First, companies should measure outcomes at the level of individual business functions. This can make the impact of AI investments easier to identify.
Second, firms must build governance into their AI systems from the beginning. Data controls, monitoring, and clear accountability will become essential as systems gain more autonomy.
Third, businesses should redesign workflows rather than simply add AI tools to existing processes.
This final point could prove especially important. Automating an existing task may improve efficiency, but redesigning a process around AI could create broader operational gains.
For Singapore's fintech ecosystem, the next stage is therefore not simply about increasing AI adoption.
The real test is whether companies can turn widespread deployment into measurable growth while maintaining the trust and safeguards that financial services require.
With 94% of respondents planning higher AI budgets in 2027, investment momentum remains strong. However, the gap between expected returns and actual results shows that spending alone will not guarantee success.
The companies that combine effective governance, measurable outcomes and redesigned operations may be best positioned to capture AI's wider commercial potential.


Comments
No comments yet. Be the first to share your thoughts.
Leave a comment