# NiCE launched CXone Agents with AI Studio creation and CXone data-layer access NiCE announced CXone Agents on June 17, 2025, describing no-code agents that connect CXone APIs, knowledge, channels and models across service workflows. Canonical: https://rangoon.ai/insights/archive-nice-cxone-agents-june-2025/ Author: Rangoon Editorial (https://rangoon.ai/insights/editorial/) Published: 2026-10-24 Updated: 2026-10-24 Event date: 2025-06-17 Tags: [NiCE](https://rangoon.ai/insights/tags/nice/), [contact center](https://rangoon.ai/insights/tags/contact-center/), [AI agents](https://rangoon.ai/insights/tags/ai-agents/), [Enterprise AI](https://rangoon.ai/insights/tags/enterprise-ai/) Source note: Historical reporting: NiCE announced CXone Agents on June 17, 2025. The release identifies AI Studio, proprietary CX AI models, APIs, Knowledge, Experience Memory, Channels and Enlighten Models. A product article first published June 18, 2025 and updated July 23, 2026 supplies current technical context only; it cannot establish the June 2025 state. Capability and outcome language is supplier reporting, not independent validation. Image: A CX request moves from channel through CXone context to fulfillment across front, middle and back office. Original Rangoon explanatory diagram: Source boundary: dated primary release reporting; original editorial analysis. No vendor interface, photograph, measurement result, or endorsement implied. ## Key takeaways - NiCE described CXone Agents as spanning self-service, mid-office approvals and back-end fulfillment, not only conversational answers. - The stated CXone substrate included APIs, Knowledge, Experience Memory, Channels and Enlighten Models, with Copilot for employees and Autopilot for customers. - A product article updated July 23, 2026 describes natural-language creation and CXone-wide deployment, but it is not proof those details were present on June 17, 2025. ## The launch described a service workflow that crossed three operational layers [source 1](https://www.nice.com/press-releases/nice-launches-cxone-agents-enterprise-grade-agentic-ai-agents-built-for-cx-to-deliver-automated-fulfillment) NiCE announced CXone Agents on June 17, 2025 as AI agents for the customer service ecosystem. The release’s scope was explicit: self-service at the front, approvals in the mid-office and back-end fulfillment. It said agents could move across systems, trigger actions and work alongside people and other AI agents. That differs from a release about summarization or suggested responses because the stated endpoint is a complete service journey. NiCE gave claims processing and refund approval as examples, which identify possible customer-impacting workflows without proving an enterprise has activated them. The announcement also says agents use proprietary CX AI models trained on use-case-specific data and optimized workflows based on top employee actions. That is NiCE’s architecture description and performance positioning, not a published independent benchmark or reusable model specification. ## The release named the CXone systems and the two deployment audiences [source 1](https://www.nice.com/press-releases/nice-launches-cxone-agents-enterprise-grade-agentic-ai-agents-built-for-cx-to-deliver-automated-fulfillment) NiCE said CXone Agents tap the CXone ecosystem through APIs, Knowledge, Experience Memory, Channels and Enlighten Models. It also separated deployment through CXone Copilot, which supports employees, from CXone Autopilot, which serves customers directly. Those names make the launch more specific than a general assertion that an agent has context. APIs provide an application boundary; knowledge and Experience Memory are described product context layers; Channels determine where an interaction appears; and the model layer informs how the platform frames automation. The public release does not expose an enterprise’s connected endpoints, retention settings or data-selection rules. It therefore cannot prove which customer data an agent could read, whether an API action succeeded, or whether a customer interaction came through Copilot or Autopilot. ## The dated release specified outcome-based prompts; later copy adds current authoring detail [source 1](https://www.nice.com/press-releases/nice-launches-cxone-agents-enterprise-grade-agentic-ai-agents-built-for-cx-to-deliver-automated-fulfillment) [source 2](https://www.nice.com/blog/reimagine-automation-with-cxone-agents-that-act-resolve-and-scale-instantly) The June 17 release says CXone identifies high-impact automation opportunities across front, middle and back offices, then can generate agents in AI Studio from outcome-based no-code prompts. That is the source-zero statement about the historical launch. A separate NiCE product article displays first publication on June 18, 2025 and an update on July 23, 2026. Its current text says a CX leader can use natural-language descriptions and conversational tools, after a specialized copilot analyzes unstructured interactions and raw data. It also says newly created agents can be deployed across the CXone ecosystem with one click, including Copilot and Autopilot. Those details may describe the product now, but without an archived June 2025 version they do not prove the June 17 release state. The materials do not publish a customer’s approval tree, test cases or a record of a generated agent. ## A 2026-updated product page describes observability and use cases, not the launch state [source 2](https://www.nice.com/blog/reimagine-automation-with-cxone-agents-that-act-resolve-and-scale-instantly) The updated NiCE product article says the platform has observability across AI models, solutions and use cases so organizations can iterate and refine agents. It also describes agents executing tasks, updating records and using knowledge across organizational systems. Its examples include replacement credit cards, loan applications, payroll adjustments, benefits enrollment, refund requests and inventory inquiries. Those examples reveal the breadth of NiCE’s current workflow framing, but they are not evidence that every detail appeared in public material on June 17, 2025. They remain supplier-supplied possible applications, not a declaration that each workflow is generally deployable without sector-specific review. The original event record supports a June launch and named architecture. The updated page cannot establish the historical timing of observability, creation flow, customer consent, approval authority or a fulfillment result for another company. ## Sources 1. [NiCE: CXone Agents launch](https://www.nice.com/press-releases/nice-launches-cxone-agents-enterprise-grade-agentic-ai-agents-built-for-cx-to-deliver-automated-fulfillment) 2. [NiCE: CXone Agents product explanation (first published June 18, 2025; updated July 23, 2026)](https://www.nice.com/blog/reimagine-automation-with-cxone-agents-that-act-resolve-and-scale-instantly) ## Related reading - [Zendesk’s 2025 Resolution Platform joined service agents, knowledge, actions and QA](https://rangoon.ai/insights/archive-zendesk-resolution-platform-march-2025/) - [Writer launched AI HQ with a beta Agent Builder and observability tools](https://rangoon.ai/insights/archive-writer-ai-hq-enterprise-agents-april-2025/) - [Workday announced Build with a future Agent Builder availability window](https://rangoon.ai/insights/archive-workday-build-agent-builder-september-2025/) ## More information - [Documentation index](https://rangoon.ai/llms.txt) - [AI access and policies](https://rangoon.ai/ai/) - [Source repository](https://github.com/hypler-dev/rangoon)