As the Federal government is looking to recover and make sense of the rampant unemployment insurance (UI) fraud seen during the COVID-19 pandemic, experts today told members of Congress that the government is in dire need of a technology overhaul to keep up with fraudulent actors who are leveraging emerging technologies such as AI to steal taxpayer money.
During today’s House Ways and Means Oversight Subcommittee hearing, experts from across industry and government said that identity theft is a real problem when it comes to UI funding, but the root cause is data.
“We have a data problem in government,” said Linda Miller, founder and CEO of Audient Group. “The use of data in government is broken. Data is an essential tool in the fight against fraud and sharing data and using it to prevent fraud is simply not working.”
“We are very protective of the privacy of citizen data – and I’m not saying we shouldn’t be – but today, when every single one of our identities are being stolen on the dark web, the government is protecting our information to the degree that the adversaries can use the data, but the government can’t use it to stop them,” Miller added.
Miller explained that every single American’s information is for sale on the dark web today, with social security numbers going for 25 cents a piece. She stressed that both at the Federal and state government level, agencies need to leverage technology – such as device geolocation – to know whether or not someone who’s applying for a benefit is using a stolen identity.
However, as Chairman David Schweikert, R-Ariz., pointed out, he could easily bust that system by using a VPN (virtual private network).
“You can,” Miller agreed, adding, “Most of these tools to effectively use stolen identities can be circumvented, especially in the age of generative AI. So, now we’ve even got much more sophisticated tools than we did the beginning of the pandemic that they’re using and exploiting right now.”
“We’ve got a dynamic adversary, and we have very, very static processes that address them,” she warned. “Agencies need to be able to use the kinds of tools that banks use, that technology companies use, where they can very quickly identify these patterns – and that is a data problem.”
What’s more, Robert Asaro-Angelo, commissioner of the New Jersey Department of Labor and Workforce Development, informed members of Congress that even if the government agency is aware that a fraud actor is applying for benefits, they are still required to let them apply.
“Right now, under statute, we have to let everybody apply for UI benefits no matter if we know that they’re stolen in advance,” Asaro-Angelo said. “This is where I think we need to marry technology and policy.”
Miller stressed the importance of changing the current statute to prevent fraudulent actors, saying “this is fraud that’s funding our adversaries.”
“There is a full underground market of fraud actors today – we call it fraud-as-a-service – and those fraud actors have at their disposal really sophisticated artificial intelligence tools,” Miller said. “We cannot fight this adversary the way that we are currently fighting this adversary because this adversary is serious, and they’re looking to steal money to create not just a financial problem but a national security problem.”
In order to better combat UI fraud, Miller called for the creation of a dedicated anti-fraud office with senior-level authority. She explained agencies struggle with a lack of incentives to allocate resources to fraud activities and said this kind of office could “affect real change” and assist with fraud data analytics.
“Such an office would have the necessary skills and resources to work solely on addressing the data accountability and technology challenges agencies face at every level of government,” Miller said.
