ING is rolling out an artificial intelligence (AI) tool, Katana, designed inhouse to help traders decide what price to quote when a client wants to buy or sell a bond.
Katana learns from the history of hundreds of thousands of trades and translates this into a prediction or suggested decision for the trader.
The first results of testing the system with the Emerging Markets (EM) desk in London demonstrated faster and more accurate pricing decisions and a concomitant 25% reduction in trading costs.
By providing traders with a visualisation of relevant historic and real-time data, the AI agent is designed to complement a trader’s cognitive ability, rather than replace him or her, says Santiago Braje, global head of credit trading at ING Wholesale Bank.
“We have seen some examples of applied AI on the trading floor before, but in most cases they have a narrow focus on automation, which restricts them to liquid asset classes where the trader is replaced by a machine,” he says. “With Katana, AI is applied to enhance the traders’ decision-making abilities, allowing them to deploy their natural intuition and expertise in the most effective way. This is a powerful combination. Our traders stay fully in control, but are now better equipped to give the best quote to our clients every time.”