Ant International Recruits Banking Giants for Updated AI Model

Ant International, a Singapore-based FinTech company, has recruited several of the world's largest banks to utilize its newly developed AI model. The company's latest creation, the Falcon Time-Series Transformer (TST) 2.0 artificial intelligence model, is designed to deliver more precise forecasting for real-world foreign exchange (FX) risk management in cross-border payments.

Ant International explained that while large language models excel at learning relationships in text, TST models are particularly crucial in finance, where liquidity needs, foreign-exchange movements, and transaction flows can change rapidly. The financial information utilized by TST models consists of continuously changing numerical data, such as transaction amounts, account balances, settlement flows, and currency positions, which directly impact capital efficiency for global payment institutions.

The AI prediction value goes beyond providing accurate calculations; it helps businesses recognize when they require funds, the amount needed, and in which currencies. Currently, banks such as Barclays, Citi, Deutsche Bank, and Standard Chartered are using the new tool. Looking forward, Ant International plans to expand the model's applications to other industries, including demand forecasting for supply chain management in eCommerce and predictive operations management in the aviation sector.

Jiang-Ming Yang, Ant International's chief innovation officer, emphasized that the company views AI value not just in achieving better forecasting scores but in turning predictive intelligence into real decisions, such as preparing liquidity, managing FX exposure, and allocating capital more efficiently. The company's AI advancement represents a significant step towards making predictive AI a foundational capability for global businesses in payments, accounts, and broader financial services.