Business and IT leaders report AI agents are scaling faster than their guardrails

The race to build an artificial intelligence agent “workforce” is on, and winning might not be about moving first, but about putting safety first.¹ It might feel like a worthwhile shortcut to skip designing and enforcing guardrails upfront in favor of adoption and experimentation, but it’s likely a slower, costlier route to try to retrofit oversight later—not to mention the potential risks to the brand and bottom line that organizations could encounter.²

According to a recent Deloitte survey of 3,235 information technology and business leaders from 24 countries across the Americas, Asia Pacific, Europe, and the Middle East—all of whom have direct involvement in their organizations’ AI programs—only 21% of respondents say their organizations have a mature governance model in place for agentic AI. The survey findings, published in Deloitte’s 2026 State of AI in the Enterprise report, reveal that agentic AI usage is scaling quickly among respondent organizations. By 2027, 74% of respondents expect their companies to be using AI agents at least “moderately.” Of those respondents, 23% expect to use it “extensively,” and 5% expect to fully integrate agents as a core component of their business operations (figure 1). Yet approximately 80% of the organizations surveyed currently lack mature governance capabilities for agentic AI, such as clear boundaries for agents that define which decisions they can make independently versus which require human approval, real-time monitoring systems that track agent behavior and flag anomalies, and audit trails that capture the full chain of agent actions to help ensure accountability and enable continuous improvement.³