PayNet Explores Federated AI Model To Bolster Fraud Detection In Digital Payments
By K.Naveen Prabu
KUALA LUMPUR, Sept 8 (Bernama) -- Payments Network Malaysia Sdn Bhd (PayNet) is exploring the development of a federated artificial intelligence (AI) model, tentatively called FinancialGPT, to enhance fraud detection and strengthen security in digital payments.
In an exclusive interview with Bernama, PayNet chief of staff Dr Endry Lim Zhen Wen said the large transaction model (LTM) would be trained solely on Malaysia’s financial transaction data to enhance resilience against cyber threats.
“We believe this model would tackle fraud and scams, ensure broader financial inclusion, and address compliance challenges specific to our market while ensuring sensitive data never leaves the country.
“It combines global best practices with national sovereignty, making it uniquely suited to Malaysia’s needs,” he said.
Lim also said PayNet is still at the early stages of this effort and is building its capabilities step by step.
“We intend for every model that we explore to undergo rigorous checks for explainability, bias and fairness because accountability is non-negotiable.
“PayNet is not doing this alone. It is actively collaborating with leading researchers at respected global institutions such as MIT CSAIL and Imperial College London on frontier issues including cyber risk, data privacy and AI governance,” he said.
These partnerships provide access to advanced research while ensuring PayNet’s approach remains aligned with Malaysia’s context and regulatory framework.
Lim highlighted that Malaysia’s push to develop as an AI hub depends on two key challenges — building talent and fostering public trust.
“The industry faces a shortage of skilled AI and data science professionals, particularly those with financial services experience, and this limits the deployment of secure, innovative and fair AI solutions at scale.
“Whereas, public trust depends on transparency and accountability, especially in finance, where customer protection and system resilience are vital,” he said.
Lim emphasised that regulation alone is not enough to meet these challenges, and genuine progress requires strong collaboration across government, industry and academia.
He also said that Malaysia’s role in driving the ASEAN Digital Economy Framework Agreement (DEFA) places PayNet at a pivotal point to help transform cross-border digital payments into a seamless, inclusive, and secure ecosystem.
“DEFA’s vision for a real-time, multi-currency payment system promises to reduce transaction costs, accelerate settlements and enhance overall payment efficiency across the region.
“This will not only drive economic growth but is also projected to expand ASEAN’s digital economy to an estimated US$2 trillion by 2030,” he said (US$1=RM4.20).
-- BERNAMA