Lie detectors to combat benefit fraud
Following the success of a voice analysis pilot project to combat benefit fraud, the government has announced more funding.
Funding of 1.5 million will be made available for local authorities to test new lie detector technology, which has already been used successfully to catch benefit fraudsters.
The Department for Work & Pensions (DWP) has announced that the money will be available to 15 local authority projects after initial results from seven pilots across 14 local authorities were positive.
The Voice Risk Analysis technology works by detecting changes in a caller's voice to detect whether the call is suspect and whether it is suitable to take any further action. Operators are trained in intelligent questioning and forms of behavioural analysis.
Harrow Council worked with Capita in testing the technology during May 2007, when it was used to carry out phone reviews of Housing Benefit and Council Tax benefit claims.
The intention was to reduce the intrusion of the review process, which was traditionally postal and visit driven. Low risk cases would simply have to forward verification of any changes, while high risk cases would be subject to further action with home visits by a fraud investigation officer.
The BBC reported that it saved 420,000, partly because claimants realised their answers were not adding up and did not want to be prosecuted for fraud. Others admitted that their circumstances had changed and were not eligible for benefit anymore.
"Harrow Council is one of seven local authority pilots nationwide, who have already recognised that this cutting edge technology can be used to stop criminals and is a unique weapon in the fight against benefit fraud," said DWP anti-fraud minister James Plaskitt.
Get the ITPro. daily newsletter
Receive our latest news, industry updates, featured resources and more. Sign up today to receive our FREE report on AI cyber crime & security - newly updated for 2024.
At the beginning of a call, the characteristics of a customer's voice frequency are sampled to establish a benchmark, which is used to analyse any changes in frequency.
The DWP said that the technology performs thousands of mathematical calculations, which results in the identification of different emotional content. It said that patterns associated with the emotional categories can identify genuine callers.
Plaskitt said: "The positive and encouraging news from the pilots shows that this technology is helping to combat benefit fraud. It is making it quicker and easier to review claims, especially for those people who are genuinely entitled to benefits."
"Overall the huge majority of people who receive benefits are entitled to them. However there is a minority who still try to steal money fro those who are most vulnerable.
He added: "We need to continue to do more to make sure that the taxpayers' money always goes to those who need it most."