VA is increasingly looking to AI to enhance claims processing
Veterans earn healthcare, education, pension, disability and other benefits from their military service, with VA personnel processing these claims as they are submitted to the department — a task that can take weeks or months to complete.
VA’s latest AI strategy, which was released in October, said additional uses of the technologies would enhance claims processing because “AI will be capable of further automating tasks such as document intake, classification, and preliminary adjudication, making it more feasible than ever for VA to deliver benefits in ‘minutes not months.’”
The department has already been adopting and piloting uses of AI across its operations, including using high-powered algorithms to help clinicians identify veterans at high risk of self-harm and rolling out an AI-assisted colonoscopy tool.
VA’s 2025 inventory of AI use cases, which was released in late January, detailed 367 examples of the emerging capabilities. Twenty-eight of these use cases are focused on the topic of “government benefits processing,” with the majority of these examples listed as still being in the pre-deployment phase.
Passage of the PACT Act, which was signed into law by former President Joe Biden in August 2022, also provided veterans exposed to burn pits and other toxic chemicals during their military service with expanded access to health services and benefits for medical conditions that were not previously covered by VA.
That law, in part, granted the VA secretary the authority “to use appropriations to enhance claims processing capacity and automation” to deal with the influx of new claims. A senior Biden official told reporters in 2024 that the department was investing in automation to help streamline its PACT Act-related processing efforts.
VA Press Secretary Pete Kasperowicz told Nextgov/FCW the department is “using AI in several ways to process PACT Act claims more quickly and efficiently.”
Some of these active AI use cases include Toxic Exposure Risk Activity Memo Automation and Automated Decision Support, or ADS.
The PACT Act requires veterans to submit detailed information about their exposure to toxic chemicals — known as TERA memos — but Kasperowicz said this process could take anywhere from 30 to 60 minutes to manually complete.
“By using machine learning, VA can now prepopulate portions of these draft memos automatically, allowing VA to process more claims,” he said, adding that it streamlines the process “but does not make any decisions about the claims.”
VA’s 2025 AI use case inventory also said ADS uses machine learning to automate “some of the up-front time-consuming development activities of retrieving information” and noted that the tool “is not intended to replace trained claims processors — it provides tools to assist with development tasks at a time when [the Veterans Benefits Administration] is receiving more claims than ever before.”
Kasperowicz said VA “plans to broaden the ADS program to cover more types of conditions and claims,” and that “this means a greater percentage of claims will be eligible for automated processing, speeding up the process and reducing the workload for claims processors.”
Addressing the claims backlog
Under VA Secretary Doug Collins, the department has touted enhanced claims processing productivity, due, in part, to the use of new technologies and automation, as well as a greater emphasis on clearing out the claims backlog.
Kasperowicz said VA reduced its backlog of disability claims by 61% during the last year, an effort that “was achieved thanks to VA leadership’s focus on workload management and offering overtime to VA claims workers to ensure the backlog is reduced.”
Some of the claims-focused AI use cases still in pre-deployment — which means the tool “is in a development or acquisition status” — are specifically focused on addressing the claims backlog.
These include an automated ratings summarization tool, which the department said “is intended to reduce the amount of time a claim processor spends sorting through documentation relevant to a Veteran's claimed issue.”
Another use case, called AICES, is an agentic AI tool that “indexes structured, semi-structured, and unstructured Veteran health and service record data, including diagnosis, severity, and service connection evidence from eFolders, metadata, and lay evidence.”
VA said AICES would help reduce the claims backlog and processing times “by more efficiently using Acceptable Clinical Evidence (ACE), minimizing unnecessary in-person exams and addressing workflow inefficiencies.”
Although VA is still looking to deploy additional AI tools to speed up its processing of veterans’ claims, the department has said its focus on improving turnaround times is already paying off.
VA announced on Feb. 23 that it had reduced its overall backlog of disability compensation and pension benefit claims to under 100,000, which it said was the first time the figure has been under six digits since May 2020. This was in addition to processing over 3 million disability and pension benefit claims during fiscal year 2025.
“VA’s claims processing productivity is the highest it has ever been, and we look forward to continuing to provide record levels of service to Veterans and VA beneficiaries,” Collins said at the time.
]]>