ISG ENI

What we do

Building the Autonomous Agentic AI Network Engineering Toolkit

ISG ENI develops use cases, requirements, and specifications that form the foundation for autonomic networks. Our work translates abstract AI concepts into concrete, implementable standards.

• We Define the Architecture:  We specify a Normative Model-Driven Architecture for context-aware, policy-based network management. This framework acts as a metasystem that sits above existing network functions, enhancing them with AI-driven reasoning without requiring a complete infrastructure overhaul.
• The Architecture is summarised as two control loops using AI modelling. Data gathered is passed via an optional API, normalised and processed in several AI Analysis Functional Blocks, which may be recursive and interactive using an inner loop. The Actionable decision is then de-normalised and passed back to the network using the same optional API as automated output. This is then monitored.
• We Enable Control and Assurance:  We produce specifications for closed-loop operational models. This includes defining how networks collect data, analyse experiential metrics, and enact changes via policy management to maintain stability and optimise performance, even in the face of unexpected events.
• We Future-Proof the Network:  By working on critical interfaces like the AI Agent protocols for 6G, we are ensuring that future standards are born with automation in mind. We are not just solving today’s operational headaches; we are laying the groundwork for the fully autonomous, intent-based networks of the next decade.
• A focused topic stream is on the AI Agents customisation, adaptation and Agentic AI design., ETSI (from ISG ENI) has published seven deliverables: AI Agents for Network Slicing; AI Agents for Core Networks and Multi AI Agents frameworks. ISG ENI is working on: a specification of AI Agent Interface and Protocols and a report on Model Training for Network AI Agents. These deliverables may be revised, Security of Agentic AI is a subject in consideration, ISG ENI will co-operate with ETSI’s Technical Committee on Securing Artificial Intelligence (TC SAI) on security.
• ISG ENI has focused on improving the operator experience, using closed-loop AI mechanisms based on context-aware, metadata-driven policies. This enables the ENI system to recognise and incorporate new and changed knowledge and hence make actionable decisions. This model gives recommendations to decision-making systems, such as network control and will interact with management systems, to adjust services and resources offered based on changes in user needs, environmental conditions and business goals.
• ENI has published a System Architecture with Context Aware Policy Management, Categorisation on Networks using AI Intent aware network Architecture, Data mechanisms, Evaluation of Categorisation, functional concepts, Prominent control loop Architectures & Artificial intelligent mechanisms. A Proof of Concept (PoC) Framework Use Cases, Requirements and Terminology in Release 4.

Explore our Work Programme