Release date: 15 September 2025

Review the top features in this release and view other new features and exciting enhancements for Automation 360 v.38 release.

  • Build 40209: Automation 360 (Cloud)
  • Build 40179: Automation 360 (On-Premises) and IQ Bot (On-Premises)
  • Build 40170: Automation 360 (Cloud-Sandbox)

We have released the updated Cloud Build 40209 (previously Build 40170), Cloud-Sandbox Build 40170 (previously Build 40011), and On-Premises Build 40179 (previously Build 40054). For information about the updates, see Release updates.

Highlights of this release

Here are the top features in the v.38 release:

Autonomous AI Agents
Autonomous AI Agents represent a shift in enterprise automation, advancing beyond rule-based, task-specific bots to intelligent, goal-driven systems that can plan, reason, and dynamically orchestrate workflows. Autonomous AI Agents use our industry-first Process Reasoning Engine (PRE).
  • Real-time adaptation: Adjust intelligently to changing contexts using structured prompts and contextual variables.
  • Tool calling at scale: Reuse existing customer automations, including bots, APIs, and other AI Agents, along with tools from third-party applications and enterprise systems.
  • Human-in-the-loop controls: Add checkpoints for approvals, escalations, and exception handling across complex processes.
  • Enterprise-grade governance: Equip developers with detailed execution logs, testing tools, and override controls for reliable deployment.

Delivered with enterprise governance and comprehensive audit trails out-of-the-box, these AI Agents uniquely combine AI autonomy with enterprise compliance.

Generative AI models from Automation Anywhere: Enterprise AI adoption is typically a time-consuming process due to lengthy security reviews that often slow down or even block generative AI deployments. In this version, we are providing pre-approved, hosted generative AI models directly within your Control Room for faster generative AI deployments.
  • Instant adoption: Use trusted models on day one with no additional integrations required.
  • Enterprise-grade compliance: All models undergo rigorous internal security and governance reviews.
  • Secure run-time: Models run within Automation Anywhere-managed Cloud environments.
  • Seamless integration: Apply models immediately to AI Skills and AI Agent reasoning steps.

Example: In healthcare, organizations can now build AI Agents for prior authorization approvals using these pre-approved models, thereby shortening compliance cycles and accelerating deployment timelines.

Note: These generative AI models are available only to public cloud (AWS) customers.

AI Agent

See AI Agent.

Agentic orchestration
AI Agents in Process Composer: AI Agents are now fully embedded into the Process Composer, enabling organizations to visually orchestrate agents, bots, APIs, and document workflows within a single, intelligent process environment.
  • Multi-agent orchestration: Design workflows where multiple AI Agents, bots, and APIs collaborate seamlessly to drive mission-critical operations.
  • Contextual next-step suggestions: PRE-powered recommendations guide builders with intelligent next actions, reducing design complexity and build time.

    Suggest next actions

  • Drag-and-drop simplicity: AI Agents are now first-class components within Process Composer, orchestrated just like bots and APIs.

    Add AI Agent to a process automation

Example: In insurance, claims processing can be accelerated by orchestrating agents for eligibility verification, policy validation, and payment approvals into one governed workflow, thus reducing claim cycle times by up to 50%.


AI Agent in Process composer
Developer productivity and experience: Scaling agentic process automation or APA solutions often slows down due to manual setup, testing hurdles, and integration constraints. Our new features streamline development, boost accuracy, and expand automation coverage.
  • Co-Pilot for Automators in Process Composer: Build workflows up to three times faster using natural language prompts powered by PRE, with best practices (naming conventions, variable setup, exception handling) applied automatically. See Build automation via prompt.
  • End-to-end queue testing in editor: Validate complete workload automations directly in the build environment, simulating live queues and ensuring business logic works as intended.

These enhancements bring enterprise-grade productivity to both Professional Developers and Citizen Developers by enabling teams to deliver faster, safer, and more precise automations that scale to mission-critical use cases.

High-scale Cloud Extraction Service
Industries that depend on document-heavy workflows often struggle with manual review bottlenecks and local infrastructure constraints. The Cloud Extraction Service delivers high-volume document processing through a secure, distributed cloud infrastructure.
  • Massive scale: Efficiently process large volumes of documents, such as claims, invoices, and contracts, with capacity that scales with demand.
  • Adaptive generative AI queries: Automatically adjust to layout variability across thousands of formats (for example, 1,000 plus vendor invoice templates) and dynamically apply the correct query at run-time.
  • Infrastructure-free: Offload processing from local Bot Runners and virtual machines (VMs) to a secure cloud endpoint.

Example: A global insurer can now process claims at scale without adding infrastructure or managing dozens of Bot Runners and achieve five times faster document processing speeds while cutting costs and operational overhead.

Note: The Cloud Extraction Service is available only to public cloud customers.

High-scale Cloud Extraction Service features

See Cloud Extraction Service.

Enterprise resiliency
Enterprises require automation that is reliable, resilient, and capable of extending into complex, legacy, and virtualized systems. Resiliency-focused innovations are enhanced to ensure agentic process automations remain robust, interruption-proof, and enterprise-ready.
  • Interrupt handler
    • Detects unexpected pop-ups or re-authentication prompts (for example, session timeouts and confirmation dialogs) in real-time.
    • Automatically executes configured resolutions, such as clicking OK or re-login sequences.
    • Resumes workflows seamlessly without manual intervention.

    See Interrupt handler package.

  • Expanded Citrix automation coverage
    • Object-based automation now works with Citrix apps even when window titles are inconsistent or dynamic.
    • Reduces dependency on fragile image-based or keystroke automations, delivering greater stability for legacy system use cases.

    See Citrix XenApp support.

Agent Interoperability - preview available
Note:

This feature is not generally available and is restricted to specific customers. Contact your Automation Anywhere account team for more information.

  • With the Model Context Protocol (MCP), third-party AI assistants can now safely trigger Automation Anywhere bots, processes, and API tasks. This expands automation across different agent systems. MCP support is powered by our Process Reasoning Engine (PRE). It smartly understands what the user wants and the special context of the request, making sure the right automation runs safely and effectively.
  • Using the Agent-to-Agent (A2A) protocol, Automation Anywhere AI Agents can now also talk to and activate third-party AI assistants. This allows for true two-way interoperability. It creates new opportunities for multi-agent collaboration, where AI systems can work together, share information, and extend automation beyond the enterprise's boundary.

See Agent Interoperability

Release updates

Date Update
4 December 2025 Build 40209 (Cloud) includes the following:
  • Fix for an issue where the Goto event in a process did not populate date variables on forms when used with variable overrides (Service Cloud Case ID: 02281779).
  • Fix for an issue where letters were truncated when mapping variables in the Loop action (Service Cloud Case ID: 02274202).
  • Fix for an issue where document processing failed in certain scenarios when users modified the default learning instance process to add a new step (Service Cloud Case ID: 02278218, 02276282).
  • Fix for an issue where data for a few work items were misaligned in the exported data (Service Cloud Case ID: 02282542).
21 November 2025 Build 40179 (On-Premises) includes the following:
  • Includes October 2025 and November 2025 package updates. See October 2025 and November 2025.
  • Introducing process visibility in the Monitoring dashboard. See Monitoring dashboard.
  • Fix for an issue where the progress bar was not displayed on the In-progress activity page after updating to the v.38 release (Service Cloud Case ID: 02278556, 02278186, 02279737).
  • Fix for an issue where the default device was removed after an attended Bot Runner user logged into the Control Room (Service Cloud Case ID: 02215755).
  • Fix for an issue where device pools on the Edit schedule page incorrectly displayed Windows-based device pools instead of Mac-based device pools when editing schedules created with macOS Task Bots (Service Cloud Case ID: 02232846).
  • Fix for an issue where failed activities appeared under the In progress activity page as active instead of being moved to the Historical activity page (Service Cloud Case ID: 02224014, 02226726, 02224942, 02230057).
  • Fix for an issue where successful Workload Management (WLM) deployments were incorrectly marked as failed in certain scenarios (Service Cloud Case ID: 02208934).
  • Fix for an issue when an error was displayed when an existing automation with actions such as If, IF/ELSE, or Error handler actions were adding in the Bot editor (Service Cloud Case ID: 02274172).
  • Fix for an issue where certain Automation Co-Pilot and Bot Agent messages were not displayed in the selected system language when the system and Automation Co-Pilot languages were set to French (Service Cloud Case ID: 02165746).
  • Fix for an issue where users were unable to check-out a parent bot (Service Cloud Case ID: 02274990, 02277259).
  • Fix for an issue where automations using the Copy action in the File package failed with the error "Unable to overwrite the file(s) to because the file is being used by another process" (Service Cloud Case ID: 02276567, 02277531, 02277611, 02277608, 02277813).
  • Fix for an issue where users were not able to assign the Automation Co-Pilot Business User license when creating an IdP group mapping (Service Cloud Case ID: 02278608).
  • Fix for an issue where automation failed with a pre-processing error when the bot contained large dependencies such as .png files (Service Cloud Case ID: 02274767, 02275913).
18 November 2025 Build 40170 (Cloud) includes the following:
  • Includes October 2025 and November 2025 package updates. See October 2025 and November 2025.
  • Introducing Cloud Extraction Service. See Cloud Extraction Service.
  • Introducing new consumption-based license model for Conversational automation. See Using the Conversational Automation Interactions license.
  • Introducing process visibility in the Monitoring dashboard. See Monitoring dashboard.
  • Fix for an issue where users were not able to use the Continue to next breakpoint option in the Debug tool when running bots containing Loop actions on large CSV files (Service Cloud Case ID: 02213300, 02212393, 02256004, 02274852).
  • Fix for an issue where work item headers and values became misaligned after inserting new work items in the WLM queue, causing incorrect column ordering (Service Cloud Case ID: 02274157).
  • Fix for an issue where the default device was removed after an attended Bot Runner user logged into the Control Room (Service Cloud Case ID: 02215755).
  • Fix for an issue where automation failed with a pre-processing error when the bot contained large dependencies such as .png files (Service Cloud Case ID: 02274767, 02275913).
  • Fix for an issue where device pools on the Edit schedule page incorrectly displayed Windows-based device pools instead of Mac-based device pools when editing schedules created with macOS Task Bots (Service Cloud Case ID: 02232846).
  • Fix for an issue where the progress bar was not displayed on the In-progress activity page after updating to the v.38 release (Service Cloud Case ID: 02278556, 02278186, 02279737).
  • Fix for an issue where users were not able to assign the Automation Co-Pilot Business User license when creating an IdP group mapping (Service Cloud Case ID: 02278608).
17 November 2025 Build 40170 (Cloud-Sandbox) includes the following:
  • Includes October 2025 and November 2025 package updates. See October 2025 and November 2025.
  • Introducing Cloud Extraction Service. See Cloud Extraction Service.
  • Introducing new consumption-based license model for Conversational automation. See Using the Conversational Automation Interactions license.
  • Introducing process visibility in the Monitoring dashboard. See Monitoring dashboard.
  • Fix for an issue where users were not able to use the Continue to next breakpoint option in the Debug tool when running bots containing Loop actions on large CSV files (Service Cloud Case ID: 02213300, 02212393, 02256004, 02274852).
  • Fix for an issue where work item headers and values became misaligned after inserting new work items in the WLM queue, causing incorrect column ordering (Service Cloud Case ID: 02274157).
  • Fix for an issue where the default device was removed after an attended Bot Runner user logged into the Control Room (Service Cloud Case ID: 02215755).
  • Fix for an issue where automation failed with a pre-processing error when the bot contained large dependencies such as .png files (Service Cloud Case ID: 02274767, 02275913).
  • Fix for an issue where device pools on the Edit schedule page incorrectly displayed Windows-based device pools instead of Mac-based device pools when editing schedules created with macOS Task Bots (Service Cloud Case ID: 02232846).
  • Fix for an issue where the progress bar was not displayed on the In-progress activity page after updating to the v.38 release (Service Cloud Case ID: 02278556, 02278186, 02279737).
  • Fix for an issue where users were not able to assign the Automation Co-Pilot Business User license when creating an IdP group mapping (Service Cloud Case ID: 02278608).
12 October 2025 Build 40095 (Cloud) includes a fix for an issue where automations using the Copy action in the File package failed with the error "Unable to overwrite the file(s) to because the file is being used by another process" (Service Cloud Case ID: 02276567, 02277531, 02277611, 02277608, 02277813).
8 October 2025 Build 40084 (Cloud) release
29 September 2025 Build 40054 (On-Premises) release
15 September 2025 Build 40011 (Cloud-Sandbox) release
Note: For documentation on previous builds of this release, see the PDFs available on this page: Automation 360 Release Notes (build versions).

What's new and changed in each product

Navigate to the following pages for details on all the updates (what's new, changed, fixed, and limitations) in each product:

AI Agent Studio v.38 release
Automator AI v.38 release
Automation Workspace v.38 release
CoE Manager v.38 release
Automation Co-Pilot for Business Users v.38 release
Document Automation v.38 release