Agentic AI Moves Big in Business, But 94% Raise Concerns About Sprawl, OutSystems Survey Finds
Boston – OutSystems, the AI development platform, today released its global 2026 State of AI Development report, revealing that businesses have made a determined move from AI exploration to implementation. Almost every organization surveyed, 96%, already uses AI agents in some way, and 97% are exploring system-wide agent AI strategies. The findings show a clear shift from pilots to production as businesses embed AI into mission-critical tasks.
However, as adoption accelerates, governance struggles to keep pace. According to the report, 94% of organizations report concerns that the rise of AI increases complexity, technical debt, and security risk. However, only a small percentage of enterprises have established a central way to manage agent AI, which means that most are using agents in all different areas.
“Our approach to working with OutSystems to find an agent solution was to start with a small, well-defined project that we felt could go into production, and that could impact the business,” said Scott Finkle, VP of Technology, McConkey Auction Group. “Our main goal for this project was to build the muscle to build AI projects going forward. OutSystems and Agent Workbench will pay us big dividends as we iterate on AI implementations.”
Agentic AI represents a significant evolution from previous AI applications, capable of automating workflows, making decisions, and adapting in real time. Gartner predicts that 40% of enterprise applications will include task-specific agents by the end of 2026, highlighting the speed at which autonomous systems are being embedded in enterprise software. According to a report by OutSystems, which surveyed 1,900 global IT leaders, 49% describe their AI skills as advanced agent or expert.
Adoption maturity varies by region, with many organizations in Australia, Brazil, Germany, the Netherlands, the UK, and the US reporting moderate progress, while France is still early in its journey. Financial services and technology organizations report the highest levels of productivity.
The impact of agent AI is most evident within IT and software development, where time-to-value is easily measurable. Thirty-one percent of respondents said AI is already part of their development processes, and another 42% have embedded AI in specific stages of the software development lifecycle. As agents prove valuable in evolving environments, 52% of organizations now rely on a human-on-the-loop model, which allows systems to operate with reduced direct supervision while maintaining supervisory control.
“The transition from AI testing to measurable business results is no longer a future scenario—it’s our current reality. The findings of the State of AI Development Report reveal a fundamental shift in which building software and building AI systems have become one,” said Woodson Martin, CEO at OutSystems. “As organizations move toward an ‘agent’ model, the challenge is no longer one of discovery, but of creating a stable architectural foundation that can connect these complex intelligent systems to drive real productivity.”
Despite the momentum, infrastructure fragmentation remains a challenge. Thirty-eight percent of organizations worldwide report mixing custom and pre-built agents, creating AI stacks that are difficult to configure and secure. While 12% have used a centralized platform to manage expansion, many businesses are still experimenting with management methods that vary by team and region.
To help businesses close the control gap, OutSystems recently introduced OutSystems Agentic Systems Engineering, a new open source AI development methodology designed to help organizations build, manage, and transform managed agent systems for the enterprise.
To learn more, download the full 2026 State of AI Development report and explore how organizations are accelerating from AI testing to implementation.
Survey Methodology
To compile this report, OutSystems commissioned a third-party survey of nearly 1,900 global IT leaders across regions. The survey examined agent AI adoption rates, use cases, development methods, and management processes, as well as the obstacles organizations face as they move from pilot to production. Survey responses were collected from December 2025 to January 2026.
