82% of U.S. Government Agencies Now Use AI Agents
More than four in five federal, state, and local government agencies have adopted AI agents capable of reasoning and acting with human oversight, according to an IDC study of 118 government leaders and decision-makers released in April 2026.
The finding marks a significant shift in how government views artificial intelligence. Sixty percent of surveyed leaders believe their agencies are ahead of the private sector in AI adoption, and 83% consider AI agents essential to restructuring their organizations.
Paul Tatum, Executive Vice President for Public Sector Solutions at Salesforce, said government leaders have moved past treating AI as an experimental back-office tool. "They see it as a critical pillar of national competitiveness and service delivery," he said. "Integrating agentic AI is now mission critical."
Government Sees AI as More Transformative Than the Internet
Fifty-six percent of surveyed government leaders believe agentic AI will have greater impact than the internet and cloud computing. Fifty-one percent said it will surpass the PC, and 46% think it will be more transformative than smartphones.
Seventy-one percent of organizations plan to increase their use of agentic AI within the next year. Nearly all respondents-94%-expect AI agents to fundamentally change how their work gets done.
James McClain, acting chief technology officer for the All of Us program at the National Institutes of Health, attributed part of this confidence to shared priorities across government. "The private sector has a million different things they have to deliver on," he said. "Government has a critical mass of common use cases-data automation, process automation, customer support. That concentration creates acceleration."
Structural Changes Expected Within Five Years
The survey projects significant organizational shifts. Eighty-nine percent of public sector leaders expect humans and AI agents to work together by 2030. Seventy-four percent anticipate most employees will have AI systems reporting to them within five years.
Eighty-three percent predict agentic AI will drive major structural changes, and 74% expect new teams and departments to emerge. Some leaders foresee workforce expansion in certain areas, particularly leadership roles.
Job functions will shift dramatically. Ninety-one percent of government leaders think most employees will transition to new roles, while 92% expect existing jobs to be altered by AI.
AI Works Alongside Humans, Not Instead of Them
Agentic AI systems can plan, make decisions, and execute tasks with more independence than traditional software tools. But they don't replace human judgment-they require it.
Cassie Kozyrkov, Google's former Chief Decision Scientist, challenged how we talk about artificial intelligence. "AI doesn't think. It doesn't think for you. Where the intelligence comes from is you," she said during a keynote in August 2025. "Instead of Artificial Intelligence, it should stand for Automation Improver, or maybe even Approachable Interface."
The real risk, Kozyrkov argued, isn't what AI will do to workers. It's what it will take from them: excuses for not acting on available information.
A separate 2025 report from Zoho found that adoption rates in the Asia-Pacific region reached 54% in 2024-nearly double the global average. APAC organizations are deploying AI primarily for customer service, coding, automation, and personalization.
To learn more about how government agencies are implementing these technologies, explore resources on AI for Government and AI Agents & Automation.
Your membership also unlocks:
