It’s no surprise the government uses artificial intelligence to automate time-consuming manual tasks, but human service agencies nationwide are seeing a high return of investment after deploying smart technologies.
AI and machine learning are transforming and augmenting how human services delivers programs and citizen-facing services, according to a recent report from Deloitte. It surveyed how government entities use these types of technologies across the full life cycle of a human services case, and estimated the associated potential time and cost savings.
Human service agencies (and their clients) face a number of challenges: managing large caseloads, long wait times for government call centers, delays in service delivery and application status updates, and language barriers with non-English speakers.
Cognitive technologies are used to address these challenges and provide better, faster services. According to Deloitte, jurisdictions in the U.S. pursue AI through three major technologies benefiting both the client and caseworker:
Chatbots: For clients to schedule appointments for human services programs and address queries. For caseworkers to address inquiries and automate client follow up and documentation.
Machine learning: Predict high-risk cases, fraud detection, prioritize resources and inspections, predict and prevent delinquency, personalize service delivery and recommend clients to service providers.
Robotic Process Automation: Automate application screening, verifications and eligibility determinations.
For example, San Diego County employed RPA to sync two different systems it used for eligibility verifications. One system stored all the required documents to verify, and the other had 500 different application forms. The systems didn’t share information, so a caseworker had to open forms from one system and look for the supporting documents in the other, which meant verifying hundreds of business rules manually.
Now, when a form is open, the RPA software automatically searches, identifies and pulls up relevant documents from the other system, reducing the time it took to approve an application from 60 days to less than a week. Similarly, Oklahoma’s Department of Human Services uses machine learning software to help prioritize child welfare cases most likely to lead to child fatalities.
Deloitte also estimated the number of labor hours an agency could save by investing in AI technologies. At most, with high-level investments, human services agencies in a large mid-Western state face up to 32.5 percent time savings within five to seven years, and annual savings of $75.1 million. And in those five to seven years, Deloitte predicts AI is unlikely to cause large job losses in the human services sector.
But to make the most of AI investments, agencies should think of a job as a collaborative problem-solving effort, rather than an individual one. This way, the human defines the problems, machines help find the solution and then humans verify the solutions.