Seventy-one percent of Federal information technology (IT) decision-makers say their agency is being challenged to grow localized artificial intelligence (AI) projects beyond the pilot stage. No existing AI-ready infrastructure presents a formidable hurdle, according to From Pilots to Proficiency: Operationalizing Federal AI, a new study from MeriTalk, a public-private partnership focused on improving the outcomes of government IT.
While the majority (59 percent) have more than 10 AI pilots, less than half have taken critical steps to establish a foundation for widespread AI integration. Further, 81 percent say their agency needs help understanding what an AI-ready compute infrastructure entails. There is strong consensus for modernization progress, with a focus on edge applications, modeling/simulations, and robotic process automation.
Focus On the Edge
Eighty-five percent say the government needs to do more to embrace AI technologies on the outer reaches. Leaders call out several applications of AI at the edge as most attractive, including: AI model training, high-performance computing (HPC), intelligent wearable technology, and intelligent video analysis.
Ahead of the Curve
Agencies leading the AI race say that federated learning, a machine learning approach that computes at the devices itself using local data, is one of their top AI priorities (90 percent of leaders versus 64 percent of peers).
The AI Opportunity
Looking to the future, 90 percent of Federal IT decision-makers agree the Federal government must quickly and responsibly adopt AI to protect the United States’ technical advantage.
“Unlocking AI’s enormous potential is a challenge that our Federal government must overcome,” said Anthony Robbins, vice president of Federal at NVIDIA. “Federal agencies understand that AI can help them more effectively meet their mission objectives, but only with a firm commitment to develop it and a clear strategy to operationalize it across the Federal enterprise.”
From Pilots to Proficiency: Operationalizing Federal AI is based on an online survey of 150 Federal IT decision-makers familiar with their agency’s use of, or plans for, AI in April 2021. The study is underwritten by Dell Technologies and NVIDIA. The report has a margin of error of ±7.97 percent at a 95 percent confidence level. To download the full report, please visit https://www.meritalk.com/study/pilots-to-proficiency/.
About MeriTalk
The voice of tomorrow’s government today, MeriTalk is a public-private partnership focused on improving the outcomes of government IT. Our award-winning editorial team and world-class events and research staff produce unmatched news, analysis, and insight. The goal: more efficient, responsive, and citizen-centric government. MeriTalk connects with an audience of 160,000 Federal community contacts. For more information, visit www.MeriTalk.com or follow us on Twitter, @MeriTalk. MeriTalk is a 300Brand organization.
View source version on businesswire.com: https://www.businesswire.com/news/home/20210524005420/en/