Optro Report Warns Enterprise AI Agent Adoption Is Outpacing Governance and Accountability
Enterprises are deploying autonomous AI agents into critical business workflows faster than they are developing the governance frameworks needed to control them, according to new research from Optro.
The findings are detailed in Optro’s report, When AI Leaves the Chat and Enters the Workflow, which examines the governance challenges emerging as organisations move from conversational generative AI to autonomous systems capable of making decisions and taking actions within enterprise workflows.
The report argues that the central governance question is shifting from whether organisations can trust AI-generated outputs to how they can establish accountability when AI systems act autonomously.
According to the research, one in three organisations already uses AI in critical resilience workflows, while 30% have never tested for agentic AI failure, including loss of control or autonomous decision-making failures. Optro argues that traditional governance practices such as periodic reviews, distributed ownership and policy-based controls are not designed for systems that operate continuously and at machine speed.
The research also highlights a gap between executive confidence in AI governance and the controls actually in place. While 58% of leaders believe their governance controls are keeping pace with AI adoption, only 18% report having active risk mitigations in place.
Organisations are already reporting consequences associated with AI deployment. Over the past 12 months, 40% reported inaccurate AI outputs, 27% experienced data breaches and 26% reported regulatory action linked to AI use, according to the report.
The growing autonomy of AI systems is also creating a new identity and access management challenge. AI agents can authenticate, access enterprise systems and perform actions on behalf of organisations, effectively giving them many of the characteristics of non-human identities. Yet business teams can deploy such systems without central IT visibility.
Optro reports that 85% of organisations have integrated AI into core operations, while only 25% have comprehensive visibility into how employees are using AI.
Security concerns are also increasing. Nearly half of security decision-makers identify agentic AI as a top security concern, while nearly two-thirds of organisations reported an AI agent-related incident during the past year. Reported consequences include data exposure, operational disruption and financial losses.
Guru Sethupathy, GM of AI Governance at Optro, said the pace of agentic AI adoption is exceeding existing governance capabilities and that organisations need to rethink governance models designed around static and manual processes.
The report positions accountability as a key consideration for enterprises moving towards autonomous workflows, arguing that organisations that establish clear ownership, visibility and controls early will have greater flexibility to scale AI without waiting for incidents or regulatory intervention to force changes.
Optro’s report also includes readiness checklists for internal audit, compliance, IT and cybersecurity, and AI governance teams to help organisations assess their preparedness for agentic AI deployment.
