Only 1 in 5 organizations are prepared to move toward autonomous AI agents, Deloitte finds

An article from Dive Brief Fragmented data systems and entrenched ways of working are impeding the transition, according to the report. Published Aug. 21, 2026 By Laurel Kalser Contributor A billboard advertising an agentic AI tool is displayed in San Francisco, Calif., on June 30, 2026. To realize the full promise of AI agents, organizations will likely have to operate in a fundamentally different way, Deloitte said. Heather Diehl via Getty Images First published on hrdive.com Dive Brief: Nearly three-quarters of U.S. company leaders expect that in the next four years, about half their business processes will be redesigned or rebuilt around artificial intelligence agents, according to an Aug. 12 report from global consulting firm Deloitte. Of the 501 U.S. senior managers and C-suite-level executives surveyed this spring, 61% expect that most AI agents will be largely autonomous, with humans serving primarily in an oversight role. Yet only 1 in 5 leaders say their organization is prepared to redesign business processes to run autonomously with AI agents, the survey showed. Respondents pointed to several obstacles impeding the transition, including “poorly documented and understood processes, inconsistent and fragmented data and systems, and entrenched ways of working,” the research found. Dive Insight: AI agents have the potential to increase revenue and deliver more meaningful work for humans, but to realize their full promise, organizations will likely have to operate in a fundamentally different way, including in the way work gets done, the report said. However, rather than AI-powered automation alone, 75% of the business leaders surveyed believe there is more value to human-agent collaboration — where employees are expected to direct AI agents, review outputs, validate quality and decide when human judgment is needed, according to the findings. Leaders recognize that getting there will be disruptive, as new job roles are created and job requirements change to leverage human-agentic capabilities, the report explained. More than 4 in 10 leaders expect “a lot” to “extreme” disruption within the next 12 to 18 months. To minimize the disruption, Deloitte offered some recommendations, such as using layering (applying AI agents on top of existing processes) as a bridge, not a destination. During the layering phase, organizations can evaluate how the role of humans should evolve, who should do the work and how the work sequence should change, the report suggested. Organizations must also provide employees with hands-on time in their AI roles, trust-building exercises to help them feel more confident about partnering with AI agents, and clear expectations regarding these collaborations, the report added. Despite technology skills gaps, workforce resistance to change and system complexity, leaders increasingly see agentic AI as a tool for rethinking growth and creating new value and not just for cutting costs, according to a 2025 KPMG study. Additionally, corporate conversations about productivity are increasingly being framed through the lens of AI, with U.S. executives largely optimistic about the technology’s future impact, a study released at the end of July by the Federal Reserve Bank of St. Louis found. Given these expectations, to maintain organizational trust, HR professionals must move beyond task-level AI use toward agentic workflows, according to a July report from Culture Amp. Yet a global survey found that only about a third of HR experts said they have done so; most organizations are still using AI as a smart assistant rather than as an autonomous agent operating within bounded authority, the platform found.