UiPath Goes All Agentic: What It Means for the Future of Enterprise Automation
The company that built its name on software robots is making its biggest strategic bet yet, on AI agents that can think, decide and act. Here is what executives should understand about the shift.
For the better part of a decade, UiPath meant one thing: robotic process automation. Software robots that mimic human keystrokes, work their way through legacy applications and grind out repetitive tasks at machine speed. It was a powerful idea, and it built a company with nearly two billion dollars in annual recurring revenue.
In 2026, UiPath is telling the market, and its own customers, that robots have stopped being enough.
Inside a few months this year, the company shipped a run of announcements that add up to the biggest strategic pivot in its history. A new orchestration layer called Maestro. A tool called Agent Builder that lets people create AI agents without deep technical skills. Purpose-built agentic solutions for whole industries. And a first-of-its-kind integration that lets external coding agents, including Anthropic’s Claude Code and OpenAI’s Codex, plug straight into the UiPath platform.
The message is hard to miss. UiPath is turning itself from an RPA company into an agentic AI company, and for anyone running enterprise automation, that carries weight.
From robots to agents, and why the difference matters
The gap between a robot and an agent is organisational.
A traditional RPA robot follows a script. It does exactly what it was programmed to do, in the order it was told, every single time. When something unexpected happens, a field moves, a format changes, an exception nobody planned for shows up, the robot stops and flags a human. That is by design. Determinism is the whole point.
An AI agent works differently. It reads context. It reasons through ambiguity. Given the circumstances, it can decide which of several possible actions fits best, and it adjusts when conditions shift. Where a robot executes, an agent orchestrates.
For a CFO, the practical difference shows up like this. An RPA robot can process ten thousand invoices overnight, as long as every invoice follows the expected format. An AI agent processes those same invoices and also handles the three hundred that arrive in an odd format, from a new vendor, with line items that do not match the purchase order, resolving the discrepancies on its own and escalating only the cases that genuinely need human judgment.
That is the leap UiPath is making. The robots stay; intelligence gets layered on top of them.
The building blocks: Agent Builder, Maestro and Solutions
At the 2026 Agentic AI Summit, UiPath laid out the architecture behind its agentic future. Three components form the base.
Agent Builder, live worldwide since April 30, is a single environment for building, testing and deploying AI agents. The design goal is accessibility. It is meant for business technologists and automation specialists, not only for data scientists. An operations manager who knows a process cold but never writes code can use Agent Builder to create an agent that runs that process end to end.
Maestro is the orchestration layer, now in public preview. What makes it interesting is that it does not care who does the work: a UiPath robot, a third-party AI agent, a human, or some mix of all of them. It orchestrates across the lot. For enterprises that already own several automation and AI tools, that matters. Maestro does not push you toward consolidation. It governs the coordination.
UiPath Solutions are the most concrete expression of the strategy. Rather than generic platform features, these are industry-specific agentic workflows that fold AI agents, traditional automation, document processing and integrations into complete, governed processes.
In financial services, UiPath shipped solutions built on its WorkFusion acquisition. Pre-built AI agents now run financial-crime compliance work, alert triage, narrative reporting, escalation, the kind of thing that used to take teams of analysts reviewing thousands of cases by hand.
In healthcare, the agentic solutions target medical-records summarization, claim-denial prevention and prior authorization, three of the most labor-intensive and error-prone processes in the sector. Each one pairs AI reasoning with UiPath’s established document processing and the controls healthcare compliance demands.
In lending, credit unions such as LMCU and Suncoast are already using Maestro to orchestrate AI agents and automation across their existing lending platforms, tightening up loan quality assurance.
Retail and manufacturing get solutions for commercial pricing, merchandising and inventory management, operational work where speed and accuracy feed straight into margin.
One pattern holds across all of them. UiPath is not handing customers a general-purpose agent and leaving them to work out where it fits. It is delivering agents that already understand specific processes in specific industries.
Opening the platform: the coding-agents bet
The boldest move came on May 12, when UiPath launched UiPath for Coding Agents, the first native integration between an enterprise automation platform and outside AI coding agents.
The first release supports Claude Code from Anthropic and Codex from OpenAI, with more to come. The concept is simple, and its reach is not. Any coding agent can now build, test, deploy and govern automations inside the UiPath platform, through plain-language conversation.
In practice, a business user describes a process to a coding agent, “build an automation that pulls daily sales data from our SAP instance, reconciles it against Salesforce, flags discrepancies above five percent and routes them to the regional manager”, and the agent produces a production-ready automation inside UiPath, complete with policy enforcement, audit trails, role-based access control and runtime governance.
Supporting several coding agents instead of building a proprietary one reflects a deliberate platform philosophy. UiPath is betting that enterprises will want the choice, Claude Code in one department, Codex in another, and that the value sits in governing what the agent produces rather than in owning the agent. One team’s preferred AI tool turns out automations that flow through the same security controls, credential vaults and compliance pipelines as everything else on the platform.
For executives weighing build against buy, that shifts the math. The bottleneck in enterprise automation was never the platform. It was the supply of skilled developers to build on it. Once a coding agent can produce governed, production-grade automations through conversation, the constraint moves from talent to process identification. The question turns from “do we have the developers?” into “do we know which processes to automate next?”
Governance as the differentiator
Running underneath all of this is a theme that is easy to miss and may be the most consequential part of the strategy: governance.
Every agent, whether it was built internally in Agent Builder or generated externally by Claude Code or Codex, enters the platform through the same governed pathway. Policy enforcement, audit trails, credential management and role-based access apply the same way to all of them. That is not a setting you toggle. It is baked into the architecture.
UiPath backed the point up with AIUC-1 certification, the first enterprise automation platform to meet the emerging global standard for AI agent security and reliability. In sectors where regulatory compliance is a prerequisite rather than a nice-to-have, banking, healthcare, government, insurance, that certification answers the biggest objection to agentic AI: “How do we trust it?”
The same governance-first stance explains UiPath’s push into the public sector, where it launched on-premises agentic AI for government agencies that need air-gapped, fully controlled environments. Agents that operate entirely inside an agency’s own infrastructure, with no data crossing the perimeter. It is a pointed answer to the sovereignty and security concerns that have slowed government AI adoption worldwide.
What this means for your automation strategy
UiPath’s agentic pivot does not ask enterprises to walk away from their existing RPA investments. The traditional robots keep running. The workflows keep executing. Nothing breaks.
What moves is the ceiling. Processes that were too complex, too variable or too dependent on human judgment for traditional RPA become candidates for automation. The developer shortage that used to choke automation programs can be eased by coding agents that build governed automations through conversation. And the headache of coordinating robots, agents and people across departments can be handled by a single orchestration layer.
The executives who get the most out of this are the ones who start preparing now. Not by rushing agents into production, but by pulling out the processes that were previously too hard to automate and looking at them again through what agentic AI now makes possible.
The robots were the first chapter. The agents are the next one. UiPath is betting the company that enterprises are ready to turn the page.
Sources:
https://www.uipath.com/blog/ai/agentic-ai-summit-2026
https://www.uipath.com/newsroom/uipath-automation-suite-delivers-agentic-ai-for-public-sector