The Complete Guide to Using AI in the Financial Services Industry in Cambodia in 2025

Cambodia's financial sector is at a practical inflection point in 2025: banks and fintechs are already deploying AI to automate credit assessments, detect fraud, and power chatbots and self‑service kiosks, but stakeholders warn that unclear rules and data‑governance gaps could slow real gains - a tension AmCham explored in its June briefing on strategic AI implementation (AmCham Cambodia strategic AI implementation briefing (June 2025)).

Regionally-minded platforms also spotlight AI's role in financial inclusion and risk analytics, showing how machine learning can extend credit to customers with thin histories while tightening fraud controls (BytePlus analysis on AI for financial inclusion and risk analytics).

For Cambodian professionals and managers who need to move from idea to impact without a technical background, practical training such as Nucamp's AI Essentials for Work can build prompt‑writing and tool‑use skills that make experimentation safer and faster - think smarter decisioning without guessing at compliance (Nucamp AI Essentials for Work syllabus).

The AI industry outlook for Cambodia in 2025 is cautiously optimistic: macro headwinds - chiefly a tariff shock that helped cut AMRO's 2025 growth forecast to 4.9% from 6% in 2024 - mean financial institutions must squeeze efficiency gains from technology while managing rising credit risks, but AI offers concrete levers to do both.

Banks and fintechs are leaning into machine learning for smarter credit scoring, fraud detection, and 24/7 customer interfaces that can score borrowers with thin histories using mobile‑phone and transaction signals, a practical route to deeper financial inclusion highlighted in BytePlus's overview of AI in finance.

At the same time, AmCham's policy discussions stress that Cambodia needs clearer AI governance so experimentation doesn't outpace oversight; without nimble rules and stronger data practices, model deployments risk delays or blind spots in compliance.

The sweet spot for 2025 is therefore hybrid: deploy focused AI pilots that reduce operational cost and NPL exposure while building governance capacity - an approach that turns short‑term productivity into longer‑term resilience and competitiveness in a market racing to diversify beyond garments and low‑margin exports.

“Diversification is a core focus of the government's Pentagonal Strategy Phase I, spanning FDI, exports, and overall economic resilience.”

Key AI use cases in Cambodia's financial services are already pragmatic and customer‑facing: lenders and payment providers use machine learning for real‑time credit decisions, transaction monitoring, and fraud detection, while chatbots and voice bots in Khmer expand 24/7 access for a largely mobile‑first population; Wing Bank's work - from QR payments at small merchants to an in‑app Khmer voice bot and a real‑time decision engine that pre‑approves about two million customers for instant loans - shows how alternative data and facial recognition can power BNPL and hyper‑personalised offers (see the GSMA case study on GSMA case study: Wing Bank AI in mobile money services).

On the risk side, Cambodian banks such as Sathapana are adopting AI platforms (finbots.ai's creditX) for credit‑risk modelling and on‑demand behaviour scorecards to raise approval rates while lowering portfolio risk (Finbots.ai partnership transforming credit risk management at Sathapana Bank).

Generative AI is also reshaping bancassurance - enabling hyper‑personalised product bundles, intelligent virtual assistants, and smarter agent support - so institutions can push tailored offers to customers without ballooning operating costs, a practical route to deeper inclusion and efficiency.

“At Wing, our vision is to use digital solutions that are relevant and convenient to improve the lives of Cambodians. It is with this vision that we have built platforms like Wingmall, Wingagri, Wingmarket and Wingpoints that ensures that we provide easier and more convenient access to finance, technology and markets for millions of our customers – individuals, corporates and SMEs alike – using these digital ecosystems that we have built for them.”

The best AI for financial services in Cambodia in 2025 is less a single product and more a blended stack: cloud‑native core platforms with embedded intelligence, local‑language LLMs, and targeted machine‑learning models for credit, fraud and liquidity.

For banks that need scale and rapid productisation, enterprise solutions like Temenos - with its AI copilots, predictive engines and cloud/SaaS architecture already powering dozens of Cambodian banks - offer a pragmatic foundation for personalised pricing, automated decisioning and lower cost‑to‑income ratios (Temenos AI copilots and cloud banking platform for Cambodian banks).

For mobile‑first lenders and fintechs, locally tuned LLMs and real‑time decision engines are essential: Wing Bank's Khmer LLM, voice chatbot and alternative‑data underwriting (which helped pre‑approve roughly two million customers from a 10‑million user base) illustrate how tailored AI delivers instant loans, hyper‑personalised offers and merchant‑side BNPL without massive branches (Wing Bank Khmer LLM and real‑time decision engine case study).

Meanwhile, central‑bank and treasury functions benefit from specialised forecasting models - reflected in the National Bank's AI liquidity‑forecasting PoC - showing that time‑series ML and scenario simulation are just as critical for systemic stability as customer‑facing AI (National Bank of Cambodia AI liquidity‑forecasting proof‑of‑concept).

The clear winner for Cambodia is therefore an ecosystem approach: modular core banking, local LLMs for Khmer UX, and focused ML pipelines for credit, fraud and liquidity, all deployed with strong data governance so innovation translates into real inclusion and resilience.

“At Wing, our vision is to use digital solutions that are relevant and convenient to improve the lives of Cambodians. It is with this vision that we have built platforms like Wingmall, Wingagri, Wingmarket and Wingpoints that ensures that we provide easier and more convenient access to finance, technology and markets for millions of our customers – individuals, corporates and SMEs alike – using these digital ecosystems that we have built for them.”

Data governance and privacy are the hinge on which Cambodia's AI promise will turn: UNESCO-backed reviews and public consultations show the Ministry of Post and Telecommunications is drafting foundational laws - including Personal Data Protection and Cybersecurity - and circulating a Draft National AI Strategy for stakeholder input, but important gaps remain that matter directly to banks and fintechs (see the UNESCO report summary on Cambodia's AI governance and the public consultation announcement for the Draft National AI Strategy).

Regulators are still ironing out procurement rules, inter‑ministerial coordination and basic cybersecurity maturity, and citizens are “not routinely informed when AI systems are used in public services” - a stark reminder that a customer may not even know a Khmer chatbot or automated credit score is affecting their access to finance.

For financial institutions, this means technical pilots must be matched by clear data‑handling practices, consent and transparency so models don't scale faster than legal and ethical guardrails; regional ASEAN guidance underscores the value of soft‑law coordination while Cambodia finalizes its own playbook.

“Cambodia is showing strong commitment to responsible innovation. With the insights from this report, the country now has a clear roadmap to harness AI's potential while ensuring ethical, inclusive, and sustainable outcomes.” - Lidia Brito, UNESCO Assistant Director‑General for Social and Human Sciences

Deploying AI in Cambodia's banks and fintechs has moved from pilot demos to live operational headaches: regulatory ambiguity - notably that the National Bank of Cambodia's 2019 technology circular does not explicitly cover AI - leaves institutions building models in a grey zone even as the central bank readies revised guidelines on technology and cyber‑risk management, a dynamic AmCham flagged in its June briefing (AmCham Cambodia strategic AI implementation briefing - June 2025).

That uncertainty, paired with internal capacity gaps around data governance, vendor procurement and model validation, means rapid experiments can be slowed by compliance queues that risk turning a promising pilot into a file that collects dust.

Regionally, ASEAN's reliance on soft‑law guidance reinforces legal ambiguity rather than eliminating it, so cross‑border data processing and dispute resolution remain operational risks (NBR report: Charting ASEAN's path to AI governance).

Practically, firms must balance resilience and speed: invest in robust data controls and AI‑aware cybersecurity - such as AI‑driven monitoring for payment rails - and pair that with targeted reskilling so frontline staff and compliance teams can manage vendor models and interpret automated decisions in Khmer (AI-driven cybersecurity for payment rails - Cambodia financial services prompts and use cases).

Cambodia's regulatory picture for AI in 2025 is best described as “under construction”: as of May 2025 there is no comprehensive national AI law, but a flurry of practical steps - from the UNESCO‑backed Readiness Assessment Model (RAM) to an MPTC working group and a draft national AI strategy - aims to close that gap (Cambodia AI law RAM, MPTC working group and national AI strategy - LawGratis). Crucially, Cambodia moved on data protection too: a draft Law on Personal Data Protection (LPDP) was released on July 23, 2025, and borrows heavily from the EU's GDPR framework, spelling out subject rights (including the right to human intervention in automated decisions), obligations for controllers and processors, cross‑border rules, mandatory appointment and notification of certified data protection officers, a two‑year implementation period after promulgation, and steep penalties - administrative fines up to roughly USD 150,000 and up to 10% of annual turnover, plus possible criminal liability (Draft Law on Personal Data Protection (LPDP) summary - Hogan Lovells). The net effect for banks and fintechs in Cambodia is clear on paper: AI pilots and automated credit or fraud systems will increasingly sit alongside explicit data‑subject rights and tighter reporting duties, so governance, transparency and documented decision‑paths are quickly becoming as important as model performance.

Building durable governance and responsible‑AI programs in Cambodia means turning high‑level commitments into everyday practices that banks and fintechs can actually use: translate the UNESCO Readiness Assessment and draft national strategy into clear procurement rules, model‑validation checklists and inter‑ministerial roles so that a Khmer chatbot's loan decision is explainable and auditable to a customer who may not even know AI was used; pair that with mandatory, cross‑functional governance teams that combine compliance, data, risk and business owners alongside human‑in‑the‑loop reviewers to operationalise the “AI + human” approach that ratings firms and regional briefs recommend.

Practical moves include a published vendor‑management protocol, routine bias and red‑team testing, transparent customer notices and appeals for automated decisions, and targeted reskilling so frontline staff can interpret model outputs in Khmer.

Startups and banks should align pilots with the AmCham roadmap for strategic AI implementation and the UNESCO‑backed readiness findings while drawing on ASEAN's soft‑law guidance to harmonise standards across borders - a lightweight, staged governance program like this turns regulatory uncertainty into a competitive asset, letting institutions scale trusted AI without waiting for every law to be finalised (AmCham strategic AI implementation briefing - June 2025, UNESCO Readiness Assessment and national AI strategy workshop, ASEAN governance brief: principles and pragmatic steps).

“Cambodia is showing strong commitment to responsible innovation. With the insights from this report, the country now has a clear roadmap to harness AI's potential while ensuring ethical, inclusive, and sustainable outcomes.” - Lidia Brito, UNESCO Assistant Director‑General for Social and Human Sciences

Strategic actions for financial institutions in Cambodia should start with clear, practical alignment: tie AI pilots to the policy signals from industry forums and regulators so experiments become scalable products rather than compliance headaches - take AmCham Cambodia strategic AI implementation briefing (June 2025) as a playbook for pairing innovation with accountability and fold conference priorities - digital inclusion, cybersecurity and ESG - into project KPIs to reinforce public trust (Cambodia Banking Conference 2025 digital innovation and inclusion summary).

Concretely, banks and fintechs should: 1) launch narrow, auditable pilots with human‑in‑the‑loop review and vendor‑management checklists; 2) harden AI‑aware cybersecurity and fraud monitoring (including AI‑driven anomaly detection) to protect payment rails and customer data (AI‑driven cybersecurity and anomaly detection for payment rails use cases); 3) invest in targeted reskilling and digital‑literacy for staff and customers so automated decisions are explainable in Khmer; and 4) embed ESG and inclusion metrics - digital wage payments and gender‑sensitive product design were highlighted at the ABC conference - to ensure AI strengthens both resilience and inclusion.

These steps turn regulatory uncertainty into a competitive advantage by making trusted, auditable AI deployments the norm rather than the exception.

“Unlocking inclusive and sustainable growth requires collaboration across government, private sector, civil society, and international partners; the pentagonal strategy provides a national roadmap with industry roadmaps and platforms for financial innovation.” - H.E. Kith Sovannarith, NBC

AI's future in Cambodia's financial industry will be practical, not hypothetical: generative models and ML can speed decisions “in fractions of a second” and unlock inclusion by scoring borrowers with thin histories, but they also bring privacy, explainability and systemic‑risk questions that regulators and firms must solve together (see the BIS review: Artificial Intelligence opportunities and risks for the financial sector).

AmCham's policy conversations underscore the same point - innovation must be paired with clear oversight to avoid pilots turning into compliance bottlenecks (AmCham Cambodia briefing on strategic AI implementation in the financial sector).

The winning path for Cambodian banks, fintechs and regulators is a staged approach: narrow, auditable pilots; stronger data governance and cross‑border rules; AI‑aware cybersecurity; and targeted reskilling so decisions are explainable in Khmer.

Practical upskilling matters: programmes like Nucamp AI Essentials for Work syllabus teach prompt writing, tool use and workplace application so non‑technical managers can run safer experiments and translate models into inclusive products.

When governance, talent and technology move in step, AI can boost efficiency and inclusion without trading away financial stability.

“Diversification is a core focus of the government's Pentagonal Strategy Phase I, spanning FDI, exports, and overall economic resilience.”