NiCE has launched the next generation of NiCE Cognigy, adding tools to build, trace and govern AI agents.
The release comes as consumer-facing AI agents begin appearing in Asia, including Meta's Muse in Singapore and Japan. That is raising new questions for companies about how their systems interact with agents acting on behalf of customers.
The updated version of Cognigy is aimed at organisations that want to move beyond single-task automation and manage AI agents across broader customer journeys. It is designed to let teams create skills and personas once, then reuse them across multiple agents.
That approach is intended to reduce duplication in development work and give companies a more standardised way to deploy customer service agents at scale. The platform can also connect with external agents, including personal AI agents used by customers, through open standards.
A central part of the release is tracing, which NiCE described as full visibility into each step an AI agent takes. That includes the data and tools the agent uses, along with the decisions it makes.
The feature addresses a practical issue for large organisations that want a record of how automated systems behave, particularly when those systems are interacting with customers or other agents. As more businesses test autonomous software in service operations, oversight and auditability have become a larger part of procurement and deployment decisions.
The update also introduces AI Assist, which NiCE described as an AI companion that supports setup, testing and improvement across an agent's lifecycle. Another addition is native voice-to-voice support, in which speech-to-speech models process audio end to end.
Source link







