How multi-agent AI is transforming telco product configuration
Agentic-AI is transforming Business and Operations Support Systems (OSS/BSS) and reshaping user experiences. From automating complex workflows to configuring products faster, explore what this technology means for your operations.
Product Marketing Manager, Business and Operations Support Systems
Senior System Manager, Business and Operations Support Systems
Strategic Product Manager, Business and Operations Support Systems
Head of Development Group Artificial Intelligence, Business and Operations Support Systems
Product Marketing Manager, Business and Operations Support Systems
Senior System Manager, Business and Operations Support Systems
Strategic Product Manager, Business and Operations Support Systems
Head of Development Group Artificial Intelligence, Business and Operations Support Systems
Product Marketing Manager, Business and Operations Support Systems
Senior System Manager, Business and Operations Support Systems
Strategic Product Manager, Business and Operations Support Systems
Head of Development Group Artificial Intelligence, Business and Operations Support Systems
Make it easier to sell, deliver and get paid with Gen-AI
Artificial Intelligence (AI) is a transformative force, Generative AI (Gen-AI) takes this transformation to the next level. By generating original content, from data-driven visuals to sophisticated network visualizations and human-like interactions, Gen-AI is redefining what we can achieve in terms of efficiency, growth, and customer satisfaction.
Integrating Gen-AI into OSS/BSS isn't just an upgrade—it's a transformative overhaul of telecom operations. By automating workflows, personalizing customer interactions, and delivering real-time, actionable network insights, Gen-AI propels growth and improves customer satisfaction. Gen-AI makes it easier to sell, deliver and get paid for telecom services. An Omdia survey, commissioned by Ericsson, highlights that 46 percent of Communication Service Providers (CSPs) expect Gen-AI to impact their OSS/BSS business within the next two years. This is not merely about improving existing processes but about fundamentally changing how we approach operations.
When will GenAI have a significant impact on CSPs OSS/BSS business?
Gen-AI can enable the automation of product configuration and configure-price-quote (CPQ) processes, among others, making it easy to sell and deliver new products. Gen-AI enhances network experiences for subscribers by providing critical insights.
Streamline product configuration with Gen-AI
Managing complex product portfolios is challenging. Traditional manual configuration processes are error-prone, which can be bad for customers and bad for business. Gen-AI replaces manual configuration processes with intelligent automation, reducing errors and time-to-market Ericsson's intelligent Product Configuration Assistant is a prime example of how Gen-AI can become a natural part of OSS/BSS to great effect.
Product Configuration spans the Core Commerce and Monetization domains of OSS/BSS. Conversational AI helps with product configuration tasks and connects to Catalog Manager and Charging. This intelligent assistant adapts to different user roles, simplifying the complexity of product management and allowing you to focus on strategic initiatives.
Video: Ericsson Product Configuration Assistant
These human-like interactions allow users to specify requirements in natural language. The assistant knows the product definition process and drives the creation of business requirement documents by talking with the user for guidance. It simplifies the complex, stepwise process of configuring the new product into the technical catalog. The Gen-AI-based assistant understands the correct steps and sequence for product creation and generates the right configuration data for BSS to process. You can configure products much faster, with less effort and resources. Additionally, products are configured correctly and consistently from the start.
64% CSPs believe introducing AI will benefit catalog management (Omdia survey)
Source: Omdia Survey: Driving business growth with AI-enhanced OSS/BSS
Explore how Gen-AI powers product configuration
Let's explore how we can automate and speed up the product configuration process.
Traditionally, creating a product offer (PO) is like navigating a maze for Product and Marketing Managers. They must check if existing product offerings can be reused manually, sift through the product catalog, and build configurations for many different product and service specifications.
With Product Configuration Assistant, managers can input their requirements in natural language, and the assistant takes the wheel, browsing the catalog to check for the possibility of reuse and creating new product offerings with changes as needed. This frees marketing managers from the usual complexity in the process.
Figure 2: Gen-AI automates manual product configuration business process
To perform the task well the assistant is competent across several capabilities:
- Knowledge base: Delivering system knowledge through an intuitive conversational approach, reducing the time spent fetching information manually.
- Configuration assistance: Accessing and creating complex specifications for fast product development.
- Contextual guidance: Provides relevant information at the right time based on active tasks.
- Task-specific journeys: Guides users through predefined workflows precisely.
Agentic-AI takes Gen-AI to the next level
The main challenge we face here is breaking down complex telecom offer configuration into specialized, autonomous tasks. A multi-agent AI architecture (Agentic-AI) helps us do this, making product configuration more accurate, intuitive and automated.
Two main types of agents are at play:
- Interactive agents, maintain context, engage in ongoing conversation, and continue to improve configurations as they go.
- Automated agents execute specific tasks independently, making sure offers are valid, within the rules and configures in the best way possible.
Autonomous reasoning and acting (ReAct) agents operate in a think-decide-act-feedback loop—continuously analyzing user intent, making informed decisions, and iterating until the optimal product configuration is achieved.
Specialized agents can complete other specialized tasks, such as:
- Product knowledge agent – Keeps an up-to-date record of telecom products, services and constraints.
- User intent agent – Understands natural language questions, identifies missing details and refines the input by asking clarifying questions.
- Configuration agent – Optimizes product bundles based on pricing rules, revenue impact, and business constraints.
- User interface agent – Adapts interactions based on who the user is, personalizing experiences.
In time, the system configures offers by itself!
Make it easy to innovate faster
We’ve been thinking about how to make Gen-AI easy to use, standardized, resilient and secure. Our driving force is a comprehensive AI framework tailored for OSS/BSS. This framework enables rapid development of many AI-driven solutions while prioritizing security and privacy.
Figure 3: Facilitate rapid development and scaling of AI applications
Download the above graph as PDF
At the core of the framework lies the Ericsson Telco IT AI Engine. For Gen-AI applications, this includes:
- Large Language Model operations (LLMOps), Retrieval Augmented Generation (RAG), and orchestration and storage layers: Ensuring efficient AI task management and data handling.
- Security features: Multi-layered security approach to mitigate risks like adversarial attacks, model inversion, and data poisoning, including risk profiling, adversarial training, encryption protocols, and role-based access controls.
- Ethical considerations: Built-in guardrails to ensure AI-driven decisions comply with ethical standards and regulations. Auditing and transparency mechanisms enable the tracking of AI decision-making and ensure accountability.
- Observability: Continuously monitors performance metrics, providing automated alerts for anomalies to maintain optimal customer experience.
To succeed here we need comprehensive integration with OSS/BSS landscape, emphasizing reusability. To accelerate AI use case development, we’ve created pre-integrated telco knowledge kits. Combining these kits with telecom knowledge sources, including TM Forum (TMF) and 3rd Generation Partnership Project (3GPP) standards helps shortcut the time to implement.
This framework provides flexibility in integrating various open source and commercial foundation models. A standard UI front-end exposing Gen-AI use cases, and service delivery agents automates activities performed by operations teams.
This AI framework provides the most resilient base to scale and accelerate Gen-AI application development.
Leap forward with Gen-AI in your OSS/BSS
By integrating Gen-AI into the OSS/BSS systems with large language models, we can reduce configuration errors and get services to market quickly and easily for greater customer satisfaction. Gen-AI can streamline complex processes, offering contextual guidance, task-specific journeys, and a robust knowledge base. By leveraging advanced AI agents, the telecom industry can offer tailored solutions for customer-centric service delivery.
Gen-AI will drive efficiency and enable you to grow, innovate and create strategic advantage. With security, privacy, and ethical considerations in hand, you can confidently navigate the challenges of today and tomorrow. Are you ready to let AI solve your real problems?
Product Configuration Assistant is a Gen-AI application under Ericsson Telco IT AI Apps, a suite of 20+ cloud-native AI and Gen-AI applications to augment OSS/BSS.
Find out more about:
Telecom analytics: transform data into action
OSS/BSS evolution for successful 5G monetization
Read more
Maximize OSS/BSS impact with AI and Gen-AI
Driving AI-enhanced business growth Omdia
5G monetization to boost revenue capture