Salesforce’s 2026 Connectivity Benchmark Report reveals a fascinating trajectory in the deployment of AI agents across enterprises. While the use of these agents is extensive, with nearly 89% of organizations employing them across most teams, a startling number of these agents operate without overseeing frameworks, leading to significant governance concerns.
The report highlights a rapid escalation in agent deployment, outpacing the establishment of governance systems designed to manage them coherently. On average, companies manage 12 agents, a number expected to rise by 67% within two years. Yet, over half of these agents function in isolation, lacking connection to a central management system. The absence of comprehensive oversight systems in 54% of organizations raises profound operational concerns.
Uncoordinated AI agents bring real-world challenges. Without a centralized registry, redundant automations often occur, wasting resources and complicating data governance. Roughly 27% of enterprise APIs operate without governance, and only 49% have effective cross-application data governance. Alarmingly, 86% of IT leaders worry about added complexity over value without better integration frameworks.
Compliance is another critical concern. AI agents making unauthorized decisions expose companies to potential regulatory pitfalls, especially concerning data protection. Andrew Comstock of Salesforce emphasizes, “The true success of an agentic enterprise isn’t found in the sheer number of agents deployed, but the overall effectiveness of those agents. We need to think about how they are discovered, governed, and orchestrated to work together.”
Not isolated from architecture issues, the expansion of applications and lack of integration hinder efficient data usage. With only 27% of systems integrated, older architecture systems create silos, complicating monitoring and rollback of agent activities. The report notes that 94% of IT leaders agree that a transition to an API-driven architecture is crucial to make agents discoverable and traceable.
In responding to these challenges, Salesforce has enhanced its Agent Fabric, utilizing a centralized registry and control layer to automatically detect agents across various platforms like Amazon Bedrock and Google Vertex AI. This framework helps consolidate agents into a single governance view, aligning with recommendations for tackling these governance challenges.
Input from Deloitte Digital emphasizes shifting from mere deployment to sustainable, scale-ready integration strategies. Kurt Anderson of Deloitte states, “This is an inflection point. Enterprises must move from simply deploying agents to operationalizing them at scale through sustainable and secure integration strategies.”
The report advises enterprises to examine internal systems through focused questions. Are all existing agents accounted for, even those deployed independently by different teams? Are systems well-integrated for consistent outputs? Moreover, how does one manage incorrect agent decisions?
As the number of AI agents is predicted to surge, fostering robust governance and integration strategies will be pivotal in realizing the potential of AI agents while mitigating the associated risks. The pressing task for enterprise leaders is to prepare now to handle the next wave of AI deployment efficiently and safely.
