June 11, 2026
NiCE Rebuilds its CX Platform Around Agentic AI Instead of Tacking It On
NiCE has rebuilt its CX platform around agentic AI, treating the technology not as a feature added to its contact centre software but as the platform’s foundation.
At NiCE World 2026 in Orlando, the company said NiCE Cognigy’s agentic AI sits native at the core, and tied two further launches to it: Workforce Empowerment Suite, and NiCE Labs. The first manages human and AI agents under one operating model, while the latter is a dedicated research and prototyping unit and feeds new AI capabilities into the product roadmap.
Agentic AI Moves to the Core
For most of the contact centre’s history, software routed interactions to people. NiCE now describes a model in which AI understands intent, resolves issues on its own, engages customers before they make contact, and coordinates work across front- and back-office systems, handing to humans only where human judgment is needed. The company has given each function a named layer.
NiCE AI Agents form the execution layer, taking action across enterprise systems and resolving customer needs across voice and digital channels without unnecessary handoffs. The Agentic Engagement Plane handles orchestration, authenticating and routing work between customers, employees, enterprise AI agents, and the personal AI agents that consumers increasingly bring to interactions. Guardian AI governs the operation, monitoring AI and human actions in real time and applying compliance controls, an area where enterprises increasingly want governance enforced continuously in production rather than handled after the fact. Agentic Analytics is on top as a discovery layer, identifying what should happen next and feeding those signals back to agents, workflows, and teams.
NiCE Cognigy provides the reasoning engine behind this, while the cloud-native CXone platform supplies the enterprise foundation of scale, routing, analytics, and security. The company said the platform meets SOC 2 Type II, ISO 27001, PCI DSS, and FedRAMP standards.
One Operating Model for Human and AI Agents
The Workforce Empowerment Suite, built on NiCE CXone, gives enterprises one framework to manage, govern, and develop human employees and AI agents together.
NiCE unifies workforce management, quality, performance, compliance, and AI operations on a single platform, so that the same rules, goals, and quality standards apply whether a person or an AI agent handles an interaction. The suite includes AI-powered forecasting and scheduling, a new Copilot for Workforce Managers, and shared dashboards and coaching insights spanning both parts of the workforce. NiCE said its generative-AI quality workflows can evaluate up to 100% of interactions, with auto-summarised evaluations that surface strengths, weaknesses, and next actions.
Jeff Comstock, President of CX Product and Technology at NiCE, said every enterprise now runs a hybrid workforce of people and AI agents, and that the suite lets them “govern, coach, and scale that workforce as one so every customer gets the same experience, whether they reach a person or an AI agent.”
NiCE Labs and the Road From Research to Production
NiCE also launched NiCE Labs, a research, benchmarking, and prototyping unit built on three functions. A research and benchmarking practice evaluates models, architectures, and orchestration approaches against real-world CX scenarios before they enter the product. A prototyping and incubation function builds AI-feature prototypes inside NiCE’s agentic portfolio and feeds the ones that show enterprise value into the roadmap. An advocacy function publishes research, reference architectures, and benchmarking findings, and engages academic institutions and model providers.
Philipp Heltewig, Chief AI Officer at NiCE, said raw AI capability and enterprise CX leadership are fundamentally different, and that NiCE Labs exists to close the distance between what AI can do in a research environment and what it delivers reliably inside a complex enterprise.
Everyone Wants the Platform
Other large vendors want the same thing NiCE does, to be the system that runs a company’s AI rather than a supplier of parts for it. ServiceNow went all in on autonomous enterprise AI at its own event this year, building a governed workforce of AI specialists across IT, CRM, and employee services.
The harder question is whether enterprise enthusiasm for agentic AI is running ahead of the evidence it delivers value at scale. Gartner has forecast that more than 40% of agentic AI projects will be scrapped by the end of 2027, citing rising costs, uncertain value, and weak risk controls. NiCE’s answer is to put governance, benchmarking, and workforce control at the centre of its story rather than autonomy alone.
