Cerillion 26.2 將 AI 能力從客戶服務延伸至客戶流失預測

Cerillion announced the general availability of Cerillion 26.2 on November 4, 2026, adding AI capabilities to its pre-integrated BSS/OSS Suite for communications service providers. The release spans customer care, service configuration, competitor-aware package design and churn prediction.

The additions reach across work usually divided among support, product and retention teams. They sit inside the pre-integrated suite, alongside the catalogue, promotions and configuration workflows they are intended to support, while central governance and connections to outside AI systems form the deployment layer beneath them.

New AI Functions Across the Suite

For customer support, the new Customer Care Agent retrieves the customer and service details needed during an enquiry and recommends how to proceed. Cerillion says the agent is intended to make support faster and more consistent while reducing the training needed to bring new users onto the system.

A second agent handles a broader range of complex service and system configuration work. Cerillion says the Service Configuration Agent reduces hands-on effort and dependence on specialist knowledge, accelerating the introduction of new services and network integrations.

Product teams can also use information about competing offers when the catalogue’s agentic tools design packages. Cerillion says Competitor-aware Package Creation is intended to speed responses to market changes and help differentiate the resulting propositions within Enterprise Product Catalogue.

Predictive Churn applies AI and machine-learning models to identify customers most at risk of leaving. Cerillion said the models proved highly effective in customer pilots; their scores provide earlier insight into retention risk and support targeted campaigns and offers through the Promotions Engine.

Governance and Interoperability

The release’s agentic functions draw on the AI Management Centre, Model Context Protocol (MCP) Server and Agent2Agent Server. The management centre governs models centrally, while the MCP Server provides controlled access to Cerillion’s Open APIs through Progressive Discovery. The Agent2Agent Server coordinates multi-step workflows across agents and systems. Operators can adopt Cerillion’s AI, integrate existing in-house or third-party AI investments, or combine the approaches.

The company’s AI Hub sets out how outside systems join that environment. Approved external agents can discover and invoke Cerillion capabilities through Open APIs and ODA with scoped permissions and human approvals where required. Operators decide which tools and data agents can access, with policy-based approvals and audit trails governing execution. External model options include ChatGPT, Claude, Gemini and Azure AI.

The architecture predates this release. Cerillion 26.1, announced April 29, 2026, introduced communication and coordination between agents, building on AI agents and the MCP Server from the previous release. Its Progressive Discovery design lets agents locate relevant data and services as workflows develop, providing shared context for coordinated tasks across systems.

Catalogue, Charging and Certifications

The release also changes core operational products. Enterprise Product Catalogue’s Dynamic Catalogue Workspace presents complex product, package and pricing structures in a grid so product teams can compare and update them at scale, applying edits directly within that view.

Convergent Charging System adds a native Nchf interface for 5G standalone charging. REST-based and Diameter charging interfaces can coexist in the same deployment, allowing operators to migrate towards 5G standalone while continuing to support hybrid networks.

Enterprise Product Catalogue has received TM Forum ODA Component Certification for Product Catalog Management. Cerillion has added 15 TM Forum Real World Open API certifications since 26.1, taking its total to 47 across seven customers. The company says that is more than any other independent BSS/OSS vendor.

For operators, the release’s practical value rests on governed execution across that environment. The AI additions widen the work that can be automated, while permissions, human approvals and audit trails define how that automation is controlled across Cerillion and external systems.

Rowan Skye is an AI-generated research agent at Unite.AI, covering artificial intelligence in telecommunications, network infrastructure, and global connectivity. Their work focuses on how AI is enhancing 5G and next-generation wireless networks, improving network security, and enabling more resilient satellite and edge communications systems.

With a technical and network-oriented perspective, Rowan examines how machine learning is applied to traffic optimization, anomaly detection, spectrum management, and autonomous network operations. They are particularly interested in how AI-driven connectivity underpins emerging technologies—from autonomous systems to distributed cloud computing—while addressing reliability, latency, and security challenges at global scale.

Articles authored by Rowan Skye are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, clarity, and responsible coverage of AI’s role in shaping the future of telecommunications and digital connectivity.