The only B2B Cross-Border Payments Winner
SINGAPORE, Nov. 20, 2025 /PRNewswire/ -- XTransfer, the world's leading B2B cross-border trade payment platform, has been recognised as Highly Commended in the Best In-house Use of AI in Fraud and Financial Crime Detection at the 8th Regulation Asia Awards for Excellence 2025. Notably, XTransfer is the only B2B cross-border payment company to receive an award at this year's ceremony.
This category recognises organisations that leverage AI to detect and prevent fraud and financial crime, with an emphasis on innovation, effectiveness, and adaptability. XTransfer was recommended for TradePilot, its in-house, world's first Large Language Model (LLM) specifically designed for the foreign trade finance industry, tailored to the complexities of B2B cross-border transactions.
TradePilot addresses the challenges traditional banks face when serving SMEs, high manual anti-money laundering (AML) costs, complex B2B transaction data, and cross-jurisdictional compliance requirements. By integrating advanced AI technologies such as big data analytics, natural language processing, and graph algorithms, TradePilot achieves automated, intelligent risk identification and control throughout the entire cross-border payment process.
"We were impressed by XTransfer's innovative application of a custom-built LLM to address the specific fraud and AML challenges in B2B cross-border trade," said a judge on the Regulation Asia Awards panel. "The solution, TradePilot, leverages a unique, proprietary dataset to automate compliance for the SME sector, demonstrating XTransfer's clear vision for making global trade safer and more accessible."
"We chose to develop TradePilot in-house because existing off-the-shelf solutions on the market cannot accurately address the unique risk scenarios in B2B cross-border trade, characterised by fragmented data and complex patterns," stated Yanfang Liu, XTransfer Co-founder and CTO. "Leveraging data from serving over 700,000 SMEs, TradePilot delivers more precise, flexible, and cost-effective risk management services."
"During development, our technical team achieved key breakthroughs, including a 9.75% increase in distributed training efficiency through self-developed hybrid communication engine and a 27% reduction in memory consumption for long-text training. These advancements ensure TradePilot maintains high efficiency and stability in processing complex foreign trade documents." Liu added.
Following a successful launch in China, TradePilot is being deployed globally and localised to meet diverse regulatory requirements. In October 2025, XTransfer released TradePilot 2.0, delivering breakthroughs in multimodal data processing and vectorised retrieval to more accurately analyse transactions and process trade-related documents, significantly improving client authentication and order management efficiency.
SOURCE XTransfer
