Nevada has a large backlog of unemployment benefits appeals. It's banking on AI to help.
Nevada has been buried in a backlog of unemployment benefits appeals that began flowing into the state’s Department of Employment, Training, and Rehabilitation (DETR) during the pandemic. The state thinks artificial intelligence (AI) may be the answer to digging itself out.
Nevada has been buried in a backlog of unemployment benefits appeals that began flowing into the state’s Department of Employment, Training, and Rehabilitation (DETR) during the pandemic. The state thinks artificial intelligence (AI) may be the answer to digging itself out.
Nevada recently became the first state government to utilize a generative AI system developed by Google to analyze transcripts from unemployment appeals hearings and generate recommendations, according to a report by Gizmodo that was confirmed by the DETR. All recommendations issued from the new AI tool will be reviewed by humans, according to Christopher Sewell, director of the DETR. The project is currently underway, DETR told AC&C.
The goal of the pioneering initiative is to streamline the process and get through appeals quicker—a task that would take up to three hours for one employee could be done in as little as five minutes using the AI system, according to a report by the Nevada Independent in June.
Such time savings could have huge benefits for the DETR, which has been facing “unforeseen staffing shortages” that have had a “substantial impact” on maintaining regular operations, according to a July memo. The contract with Google AI, approved by the state’s Board of Examiners in August, cost $1 million.
To maintain privacy of those whose appeals are being processed, the contract prevents Google from accessing “personally identifiable information” that is frequently used in unemployment claims. Google is also prohibited from using confidential data in the transcripts for uses outside the contract, according to the Gizmodo report.
“The technology employed is well-established and successfully implemented in other projects and represents the cutting edge of IT development, albeit with limited prior deployments,” Lisa Jean, the Nevada’s Technology Investment Notifications administrator, stated in a memo in June.
Timothy Galluzi, Nevada’s chief information officer, has championed AI technology and its use in government operations, recently leading a State AI Roundtable to engage in discussion about the technology and how Nevada can become a leader in its usage.
Not everyone is convinced, however. Critics have raised concerns over bias and questionable accuracy in emerging AI technology, which Google itself has noted poses a risk of “unfair bias.”
Nevada State Senator Skip Daly (D-Reno) has also urged caution regarding states’ use of AI, telling the Nevada Independent that he was “dubious” of overreliance on the new tool.
“I hope that we are cautious about it, and think before we just say, ‘We got to be faster or better than the next guy,’” Daly said.
Nevada released state guidelines for the use of AI last year, stating, among other things, that such systems should undergo regular audits to maintain security and privacy.
“AI tools must be used ethically, avoiding biases and ensuring fairness and transparency in decision-making processes,” the guidelines state.
In April, the Biden administration issued its own federal AI guidelines that advocated human scrutiny when it comes to the administration of public benefits programs.
“AI should be used for business functions that are well understood and where staff have the knowledge and skills to evaluate performance,” the U.S. Department of Agriculture’s AI guidelines state. “AI should not be used for immature business functions with a goal of an AI discovering new approaches or efficiencies.”
As of this past summer, the DETR’s backlog of unemployment appeals exceeded 10,000, according to the Nevada Independent, but if all goes according to plan for Nevada, that number could dwindle faster than expected.