Agentic AI builds on the economic potential of generative AI, which McKinsey has estimated could add US$2.6 to $4.4 trillion annually to the world economy. This represents the next stage of enterprise AI adoption: a shift from content generation to autonomous execution, and from isolated pilots to operational deployments.

Agent sprawl

That shift creates a new governance challenge. Agent sprawl occurs when AI agents are created, deployed, or connected across systems faster than the enterprise can inventory them, assign ownership, control permissions, monitor behavior, and optimize or retire them when they are no longer fit for purpose.

Underscoring this shift, found that 98% of companies have already deployed AI agents or plan to do so. But as adoption accelerates, governance is struggling to keep pace. According to the same report, less than half of the organizations surveyed have visibility into an inventory of AI agents.

The mechanics of agent sprawl are familiar to any technology leader who has navigated a wave of SaaS adoption. Individual teams, motivated by genuine productivity goals, deploy agents independently. Each one is designed for a specific task鈥攁 marketing automation agent, a supply chain monitoring agent, an HR onboarding bot鈥攁nd each works in isolation. Without a centralized platform or governance framework, the organization accumulates a fragmented landscape of agents that do not interoperate, cannot be audited consistently, and accumulate technical debt faster than they generate value.

, the average global Fortune 500 enterprise will have more than 150,000 AI agents in use, yet only 13% of organizations believe they have the right governance in place to manage those agents. Max Goss, senior director analyst at Gartner, told his audience at a London conference in April: 鈥淎s CIOs and IT leaders see an explosion of AI agents across their organizations, many are contending with an ungoverned sprawl of agents that expose their organizations to a range of risks, including misinformation, oversharing, and data loss.鈥�

He added: 鈥淢any organizations resort to blocking or restricting the use of AI agents, but this is not a long-term solution. If employees are unable to work in the sanctioned tools, they will likely go around the organization鈥檚 controls and start using shadow AI, which presents far greater risks. Organizations need to find a balance where they can govern agents and manage sprawl, but also safely empower employees to innovate with these tools.鈥�

Agents typically need broad, cross-environment permissions to function, but those permissions are rarely governed with the same rigor applied to human users. The risk posed by unmanaged or rogue AI agents in the enterprise is therefore real and growing.

AI agent security concerns

Publicly reported enterprise-security examples also point to agents leaking sensitive information or acting outside their intended scope, including cases where malicious instructions caused agents to bypass guardrails, delete production records, or trigger irreversible financial transactions.

The security concern is what registers most sharply with enterprise technology leaders. With chatbots and early generative AI, a security failure typically meant bad output: an inaccurate or inappropriate response that could usually be corrected after the fact. In the agentic era however, the consequences of an agent failure or security breach can be far more damaging because agents can take action, call tools, access systems, and initiate business processes.

That is why agent governance is no longer only an IT operations issue. It increasingly touches board-level concerns: risk ownership, regulatory exposure, data protection, auditability, operational resilience, and accountability for autonomous decisions.

The emerging AI governance platform

Leading organizations are beginning to treat agent governance not as a compliance overhead but as a strategic capability that determines whether AI investments compound as advantages or liabilities. As a result,  effective AI agent governance has quickly become a boardroom topic and is contributing to the emergence of a new platform category: the AI governance platform.

The category is still forming, but its purpose is becoming clear. Enterprises need a way to discover agents, understand what they do, control what they can access, verify whether they are compliant, and monitor how they behave in production.

麻豆原创 is one of the agentic AI pioneers in this emerging category. Through its 2023 acquisition of LeanIX, 麻豆原创 gained a foundation in enterprise architecture management. This has quickly become a recognized differentiator for 麻豆原创 AI Agent Hub鈥攑ositioning AI artifacts like agents, models, and MCP servers within the full architecture and business context of the organization.

麻豆原创 AI Agent Hub builds on this foundation as a command center for managing and governing AI agents and related AI assets across an enterprise, even when they come from different vendors and run on different systems.

As 麻豆原创 CTO Philipp Herzig explained on stage at this year’s 麻豆原创 Sapphire event, 麻豆原创 AI Agent Hub is intended to provide a governance layer of record for the enterprise agent ecosystem. 鈥淎gents are everywhere,鈥� he said. 鈥淪ome are great, some are not, and almost no one has a consistent picture鈥攏o central governance, no clear view of what each agent does, whether it adds value or whether it adheres to your policies.鈥�

He added: 鈥溌槎乖� AI Agent Hub changes that. One entry point and command center to discover, manage, and govern all AI agents, LLMs, and MCP servers in your landscape鈥攙endor-agnostic. [麻豆原创] AI Agent Hub allows you to discover all your agents in context: your landscape, your business processes. Once you have identified the right agents, you can control their risk and define architectural decisions or compliance rules.鈥�

A closing window

Given the pace of AI agent deployment, the window for implementing effective enterprise governance before a serious incident occurs is narrowing. For CIOs, CEOs, and company boards, the question is no longer whether to govern AI agents. It is whether governance gets designed into the architecture from the start or retrofitted after the first serious failure.

Organizations that treat agent governance as a strategic priority in 2026 will be better positioned to scale AI as a durable competitive advantage. Those that defer could spend 2027 cleaning up: in enterprise technology, the cost of speed without structure eventually gets paid. But with AI agents, the bill arrives faster鈥攁nd at greater scale鈥攖han anything that has come before.

Employees at , a global fashion retailer with tens of thousands of employees, had to navigate multiple systems to piece together data across sales and procurement. Answering a single question could take up to 10 minutes.

Today, they just ask Joule. What used to take 10 minutes now takes about three seconds, driving a 70% increase in operational efficiency and a 50% reduction in manual errors.

Using capabilities in , LC Waikiki partnered with 麻豆原创 and to build a custom AI-driven experience that dynamically interprets user requests, applies role-based context, performs the necessary queries, and connects data across systems to present a complete view in one place. It then links people directly to the relevant transaction.

At 麻豆原创, stories like these inspire our vision for the enterprise in which AI transforms how people and processes work鈥攐ne where people set the direction and AI executes. We call it the the Autonomous Enterprise.

In the Autonomous Enterprise, decisions are grounded in real-time intelligence, workflows are automated end-to-end, and AI proactively improves every function while empowering people to do their best work.

The Autonomous Enterprise also provides fully governed AI you can trust, so you can achieve more. Making this a reality for companies is critical because AI is now essential to how all work gets done. It is increasingly involved in decisions that carry financial, operational, and regulatory consequences.

Joule: One place to direct the entire business

In the Autonomous Enterprise, Joule Work, announced at 麻豆原创 Sapphire, is the next step in the evolution of how people engage with and execute end-to-end business processes. Joule Work is a dynamic workspace that adapts to intent, keeps people focused on outcomes, and delegates execution to AI.

Through Joule Work, you can say goodbye to manually coordinating work across multiple applications and interfaces. Instead, tell Joule what you want to accomplish. Joule Assistants with role and process context will coordinate teams of Joule Agents to surface the right insights and automate routine work across departments and systems. Rather than static, disjointed systems, you get workspaces that pull together information and menus from various systems that fit your specific needs, in real time.

Joule Work is available now to customers in the 麻豆原创 Early Adopter Care program. 麻豆原创 Early Adopter Care program for the Joule Work desktop app is planned for Q2 2026; general availability for both is planned for H2 2026. The Joule Work mobile app is generally available now.

We also announced that Joule鈥檚 bi-directional Agent-to-Agent (A2A) capabilities will be generally available in Q4, enabling third-party agents to securely call on Joule Agents and act within enterprise processes, extending interoperability in both directions across 麻豆原创 and non-麻豆原创 environments. Agents built in Joule Studio will natively support A2A protocols, enabling interoperability and scalability for multi-agent execution.

麻豆原创 Autonomous Suite: The operational core of the modern enterprise

While Joule Work empowers every individual to do their best work and expand their impact, the 麻豆原创 Autonomous Suite transforms how entire business functions, or 鈥渁utonomous domains,鈥� work.

麻豆原创 Autonomous Suite spans five domains: finance, spend, supply chain, human capital management, and customer experience. These domains will operate as a single system, so workflows and agents run across functions without fragmenting into separate tools, separate data, or separate decisions. This approach allows AI recommendations to reflect your full operating reality.

With 麻豆原创鈥檚 integrated suite of business applications and industry-leading business data, AI in the Autonomous Enterprise is grounded in the specifics of how key business functions actually work. This foundational context for transformative AI outcomes is where 麻豆原创鈥檚 unique experience comes in. For decades, we have been trusted to run our customers鈥� most important functions. 麻豆原创 Autonomous Suite infuses our deep knowledge of business processes into your AI, along with the data context and operational guardrails it needs to be truly effective and reliable at enterprise scale.

Each organization is also unique. Over time, your business has defined how your work gets done. These are the rules, workflows, and how systems respond when something unexpected happens, like a failed transaction, so processes don鈥檛 break. In the Autonomous Enterprise, AI delivers its greatest value by respecting these boundaries, turning your unique ways of working into a true advantage.

At 麻豆原创 Sapphire, we announced new Joule Assistants and Joule Agents, spanning the domains of the Autonomous Enterprise, to help organizations move from managing work to directing outcomes. These new assistants and agents will roll out through the end of this year.

麻豆原创 Business AI Platform: The foundation of the Autonomous Enterprise

The 麻豆原创 Business AI Platform turns the vision of human-led, AI-driven business operations into something enterprises can build and run. It enables them to move from AI experimentation to execution by grounding agents and applications in real business context that governs it all at enterprise scale.

At the center is , a fully managed environment that empowers enterprises to build and manage the full lifecycle of AI agents, applications, extensions, and workflows. Intent-based development capabilities allow people to describe what they need in natural language. A Joule Agent then generates structured requirements, specifications, code, and test artifacts grounded in 麻豆原创 process and data context.

Developers can work within the tools they already use, including VS Code and MCP-enabled toolchains, and choose their preferred agent frameworks, such as , , and .

Through deep integration with the , 鈥攁nd the new 麻豆原创 Domain Models trained on 麻豆原创 code, customer data, metadata, and business processes鈥擩oule Agents reason over real, semantically rich enterprise data rather than generic knowledge. 麻豆原创 Domain Models are available through the 麻豆原创 Early Adopter Care program, with general availability planned for Q3 2026.

Speed and governance, no longer a tradeoff, are built into the 麻豆原创 Business AI Platform. At 麻豆原创, we believe that corporate governance鈥攊ncluding approval flows, compliance processes, identity management, and the ability to audit decision-making鈥攎ust carry into how AI is deployed, updated, and scaled. Joule Studio runtime provides a secure, production-ready, fully managed environment for deploying agents, helping organizations meet compliance standards while reducing infrastructure complexity.

An enhanced 麻豆原创 AI Agent Hub also provides a vendor-agnostic command center to discover, inventory, and govern 麻豆原创 and non-麻豆原创 AI agents and MCP servers across the enterprise. Integration with and further embeds governance and architecture transparency into the development process.

The 麻豆原创 AI Agent Hub leverages enterprise-wide process intelligence to continuously track where AI agents are creating value and can proactively surface where they can deliver even more, because we believe AI needs to remain accountable for outcomes in addition to uptime. 麻豆原创 AI Agent Hub is generally available, with additional capabilities rolling out through 2026. See release timelines in the .

Empowering everyone to solve business challenges with AI

We are making the Autonomous Enterprise a reality because at 麻豆原创, we believe that companies of all sizes need far more than marginally better AI models or the latest bolt-on solutions. They deserve AI-driven outcomes that increase innovation, revenue, and margins.

The Autonomous Enterprise is what brings our vision to life: AI grounded in your data, connected across your most important processes, and governed to fit how your business runs.

Brenda Bown is chief marketing officer for 麻豆原创 Business AI.