Fast Tests were conducted to demonstrate the potential impact of artificial intelligence in preventing cross-border payment fraud.
According to Monday (September 15), the global messaging system collaborated with 13 banks in an experiment using privacy-enhancing technology (PET) to use privacy-enhancing technology (PET) to enable agencies to safely share fraud insights across borders. press release.
In one example, the pet allowed participants to verify intelligence in questionable accounts in real time. “Developments that can speed up time to identify complex international financial crime networks and avoid fraudulent transactions being carried out,” the release states.
In another case, participants adopted a combination of pets and federated learning, or an AI model where they could “visit” the institution to train their data locally, collaborating and collaborating to find them without sharing customer information. Abnormal Transactionsevery release.
Trained using synthetic data from 10 million artificial transactions, the model was twice as effective at identifying fraud than models trained using a data set from a single institution, the release said.
“These experiments demonstrate Swift's convening power as a trustworthy cooperative at the heart of global finance.” Rachel LevySwift's AI director said in the release. “Unified industry-wide fraud defense is always stronger than what was posted by a single agency acting alone. The industry loses billions. [of dollars] Although we do scams every year, we have been able to safely share our intelligence across borders, which has significantly reduced this number and allowed us to stop scams in minutes rather than hours or days. ”
In the wake of these experiments, Swift plans to expand participation before starting the second round of testing. It states that it uses real transaction data in the hopes of demonstrating the impact of technology on real fraud.
When it comes to maintaining trust in financial transactions, Data sharing It's important.
“It's a team sport.” Entersekt Chief Product Officer Pradheep Sampath He told Pymnts in August. “And the thread that brings us all together is practical, honest and shared and responsible for controlling data.”
Subscribe daily for all PYMNTS AI coverage AI Newsletter.
