MeriTalk, a public-private partnership focused on improving the outcomes of government IT, today announced the results of a new report − “The Federal AI Journey: Insight into Present State and Opportunities Ahead” − that frames the current state of artificial intelligence (AI) in the Federal government. The study, underwritten by Pure Storage, surveyed 150 Federal IT managers and C-level Federal IT executives and identifies AI as the next frontier – with 77 percent of respondents saying it will change the way government thinks about and processes information and 61 percent saying that AI could solve one or more of the challenges their agency faces today.
New Report: AI is the next frontier – 77% of Fed IT managers say AI will change the way government thinks about & processes info & 61% say AI could solve one or more of the challenges their agency is facing. View findings here: https://bit.ly/2MfVaf7
The road to widespread AI adoption is still in the early stage of construction. According to respondents, a solid data foundation is essential to success, with 88 percent of Federal IT managers saying that data is key to better outcomes. But, nearly all respondents (97 percent) say their agency still struggles with data challenges – and 73 percent say that their challenges extend beyond harnessing and securing data to analyzing and interpreting it. In fact, only 17 percent say their agency is completely successful when it comes to their ability to draw insight from the data they collect.
While the clear majority of agencies are in the early stages of AI planning, approximately one in 10 Feds say their agency has implemented at least one AI pilot program and is already experiencing benefits. The Department of Defense (DoD) is leading the way – 83 percent of DoD managers report taking action on an AI initiative (including preparing a business case, implementing, or benefiting from a pilot program) versus 47 percent of civilian managers. More than half (56 percent) of DoD respondents indicate their agency is in the planning phase of AI while 45 percent of civilian agencies report they are in the discovery phase.
Even with persistent challenges, agencies are optimistic and are looking to ramp up AI adoption. Nearly half of Feds say AI is part of their agency’s technology roadmap, and 37 percent say they are working to add it. Agencies with a comprehensive data strategy are three times more likely than those without one to include AI as part of their technology roadmap – 61 percent versus 19 percent.
“The agencies making the most progress towards AI are those who have built or are in the process of building a solid foundation for data organization, storage, and security,” says Gary Newgaard, vice president, Public Sector, Pure Storage. “We are all chasing speed and accuracy – and if agencies can hit milestones at a faster pace, constituents will reap those benefits. The journey must start with a comprehensive data strategy and data-centric architecture – which are proven to accelerate progress and fuel AI initiative success.”
Agencies who have not yet begun using AI are nearly twice as likely as AI leaders to say fear or lack of understanding will significantly slow adoption – 62 percent versus 32 percent. For implementation to progress, agencies highlight the need for IT staff knowledge, IT infrastructure, analytics capabilities, and data organization. The other hurdle agencies are facing is complexity in the marketplace – with more than half of respondents (55 percent) saying this has made it difficult to find a solution that best fits their needs.
“AI is literally a panacea for U.S. government efficiency and beyond that for turbo-charging the U.S. and global economy,” said Steve O’Keeffe, founder, MeriTalk. “But, the government needs to move to put in place the right foundation, education, roadmap, and governance – that means organizing the gold mines locked in government data sets to unleash the full potential of this truly disruptive wave of technology. AI sits at the crossroads of cloud computing, internet of things, and cybersecurity.”
To continue down the path towards AI, the report recommends that agencies begin by embracing data centric architecture – they should focus on building a solid foundation, experimenting with AI building blocks like automation and deep learning, and take a macro view by incorporating AI into their overall technology strategy.
In this report, Federal IT managers and executives can learn how their agencies progress on the road to AI compares to others and gain valuable insight from those who are further along in the journey. Additionally, the report outlines how DoD agencies are uniquely using AI and how that differs from other agencies.