With an annual spend of more than $100 billion on IT products, the federal government is falling short on the consistent tracking of its software licenses, leading to missed opportunities for cost reductions, the GAO found. And though there are federal initiatives in place to “better position agencies to maximize cost savings when purchasing software licenses,” the GAO noted that “selected agencies have not fully determined over- or under-purchasing of their five most widely used software licenses.”

The GAO’s study looked at software licenses purchased by the 24 Chief Financial Officers Act agencies, finding that 10 vendors made up the majority of the most widely used licenses. For fiscal year 2021, Microsoft held by far the largest share of vendors organized by the highest amounts paid (31.3%), followed by Adobe (10.43%) and Salesforce (8.7%).

While the GAO was able to identify and analyze vendors based on government spend, it was “unclear which products under those licenses are most widely used because of agencies’ inconsistent and incomplete data,” the report noted. “For example, multiple software products may be bundled into a single license with a vendor, and agencies may not have usage data for each product individually.”

“Without better data, agencies also don’t know whether they have the right number of licenses for their needs,” the report continued.

For their recommendations, the GAO focused on nine agencies based on the size of their IT budgets and then zeroed in on the five most widely used licenses within those agencies. The selected agencies were the Departments of Agriculture, Energy, Housing and Urban Development, Justice, State and Veterans Affairs, as well as the Office of Personnel Management, Social Security Administration and USAID.

The recommendations centered most on better and more consistent inventory tracking to ensure that agencies didn’t double-dip on software license purchases and were in a better position to take advantage of cost-saving opportunities. There should be more concerted efforts to compare prices, the GAO stated.

HUD did not say whether it agreed or disagreed with the GAO’s recommendations, while the other eight agencies said in responses that they did.

Congress in 2023 attempted to rein in duplicative software across the government with the Strengthening Agency Management and Oversight of Software Assets Act, which aimed to consolidate federal software purchasing and give agencies greater ability to push back on restrictive software licensing. However, after passing the House in July, the bill never moved in the Senate.

The post Software license purchases need better agency tracking, GAO says appeared first on FedScoop.

During a press conference after the agency’s monthly LinkedIn Live series, #TalkingTech, Chief Information Officer Kurt DelBene and Chief Technology Officer Charles Worthington reported that the VA Office of Information Technology has “line of sight” to hire “multiple hundreds of people,” including those with AI capabilities to work with the emerging technology, but are unable to do so because of the current challenges posed by Congress’s failure to pass a long-term budget. 

While the department is currently hosting one of its tech sprints, as required by the White House’s AI executive order, DelBene said that this project is using existing infrastructure and staff, much like other projects and initiatives throughout the VA. 

“We want to staff up in the AI space in particular and that becomes challenging under a” stopgap spending bill, DelBene said during the press conference. “Obviously, the applied for budget in [2024] is larger than our existing budget and that’s what’s necessary to add more headcount. … We’re OK for now, but for long-term needs, we just need to get a budget.” 

DelBene said he thinks that the agency will leverage existing staff throughout different parts of the VA, and that its desire to bring on new talent is a “modest request.” He noted that the Veterans Health Administration is already utilizing employees to work with AI. 

“The core key is to bring it together, come up with a governance structure, come up with a prioritization of the different AI initiatives,” DelBene said. “We need that centralized group, and that’s one initiative that I think is really critical.”

During the event, Worthington said that the agency is going to try to scale its use of AI with the help of the VA’s National AI Institutes, which has been considering how the agency can responsibly adopt the emerging tech inside of the health care domain. 

“We’re going to be expanding on that work and really looking at ways to safely scale up the use of some of these novel technologies in the workflows where it makes sense,” Worthington said. “I think that there’s a lot of opportunities to use this technology to make the software work better for our staff and for veterans.”

Worthington compared the adoption of AI to the department’s transition to cloud technology and the move to make digital services available for mobile devices. He also emphasized that the agency is taking a “measured approach” and is in the process of training and upskilling staff. 

“I think in the coming year, and especially in the next few years, we’re really going to see a lot of those solutions reach scale,” Worthington said. “I think we need to be very careful on how we train our staff about how to use [AI], but with that training, I do think that there’s a lot of power that these tools can offer in a way that I’m really excited about.”

The post VA struggles to bring on AI talent without a long-term budget from Congress appeared first on FedScoop.