The Department for Work and Pensions (DWP) has shifted its approach to the use of artificial intelligence (AI) in tackling benefit fraud, according to senior civil servant Neil Couling. In evidence presented to a committee of MPs, Couling stated that the change in policy came in response to feedback from claimants and elected representatives. The DWP had deployed an AI tool, utilizing machine learning, to analyze historical data and identify higher-risk claims for investigation. However, concerns were raised about potential bias and the lack of transparency in the system. The DWP, for the first time, publicly acknowledged the alteration in its strategy, abandoning routine suspensions of claims flagged by the AI tool. Instead, claims are now only put on hold if claimants fail to respond to investigators' inquiries.
The DWP's AI tool, at the center of its plan to combat increasing levels of benefits fraud, was projected to save £1.6 billion by 2030/31. The algorithm has been used since 2022 to screen claims for Universal Credit advances and has expanded to detect fraud in areas such as income and housing cost declarations. The department, facing concerns about potential delays and unfair treatment of legitimate claimants, has not disclosed the number of claims flagged by the system. Despite not revealing the algorithm's details, the DWP claims to have processes in place to address potential bias, with promises to provide more information about its impact in the next set of annual accounts due later this year. The DWP emphasizes that a staff member will always make the final decision in determining fraud or error.
SOURCE: US NEWS | JANUARY 22 2024
IF YOU FEEL YOU HAVE BEEN A VICTIM OF FRAUD, TAKE ACTION
RIGHT AWAY. FOR QUICK INVESTIGATION ASSISTANCE,
