Major Telecom Operators Prepare for 6G: How AI Agents are Reshaping Network Architecture and Generating €200M

The telecommunications industry is on the verge of revolutionary changes. While AI was once considered an optional add-on, today it is becoming a central element of next-generation network architecture. Major operators such as Orange and NTT Docomo are already actively implementing AI agents, moving towards "AI-native" approaches where AI not only optimizes processes but completely reconfigures network logic. Orange, for example, implemented 150 AI use cases in 2024, generating €200 million in value, demonstrating the potential of AI agents in telecom.

Telecom operators deal with vast amounts of data and complex systems, where even a minor failure can cost millions. Traditional approaches, where the network reacts to problems after the fact, can no longer meet the growing demands for speed and reliability. This leads to downtime, customer loss, and missed revenue. Today, this can be changed by moving from reactive to proactive management, where the network itself anticipates and prevents problems.

The Problem: Outdated Architecture and Missed Opportunities

Modern mobile networks, particularly 5G, were primarily designed for data transmission. Their architecture is based on rigid protocols and rules, where devices request predefined services. This creates limitations: the network reacts to events but cannot anticipate them or adapt to changing conditions in real-time. As a result, operators spend vast resources on manual management, troubleshooting, and optimization, while losing potential revenue from new services.

While other industries are actively adopting AI to transform their operations, the telecommunications sector risks falling behind by focusing solely on maintaining the "data pipe" and missing opportunities to create value in client-facing and enterprise segments.

The Path to AI-Native Networks: From Overlay to Core

Recognizing these challenges, telecom operators and the 3GPP standardization body are actively preparing for 6G, where AI will become not just a tool, but a fundamental principle of network construction. The goal is to evolve from 5G's communications-centric architecture to a more cognitive and intent-driven architecture. This means AI will not merely optimize existing domains but will become the core that supervises and coordinates traffic management, services, and resources.

Companies like NTT Docomo identified AI as one of the key priorities for 6G three years ago, alongside sustainability, efficiency, and improved customer experience. Orange, in turn, scaled AI adoption in network environments, implementing 150 use cases in 2024 that generated €200 million in value. This proves that the shift to AI-native is not just an ambition, but already a reality delivering tangible results.

Designing AI Agents for the 6G Core

AI agents in the 6G Core must be capable of interpreting goals, finding resources, establishing trust, coordinating computing, and completing tasks across a range of domains: from end-users and applications to devices and the network itself. These are not just "smart" algorithms, but essentially autonomous entities that can interact with each other and with the network infrastructure based on their "intents."

A key difference from current systems lies in the evolution of the signaling layer (NAS) towards "intent-based" communication. A device or agent no longer requests a rigid Quality of Service (QoS) profile; instead, it can signal the objective it is trying to achieve. This allows the 6G AI Core to autonomously determine how best to fulfill the request, proactively coordinating connectivity, compute, security, and latency.

For example, Turkcell is actively researching intent-based communications, developing new 6G core functions to handle such requests and studying the impact of AI agents on signaling flows. Chinese operator China Mobile, within its ACN (Agent Communication Network) initiative, demonstrates how AI agents can connect various device types, including robotics, drones, AR glasses, and digital agents, ensuring trusted access, dynamic subnetworks, and multi-agent coordination.

Implementation and Standardization

The implementation of AI agents in telecommunications is happening not only at the level of individual companies but also through standardization. The role, resources, and policy access for AI agents are already being discussed within 3GPP SA2 6G on key architectural issues, including Non-Access Stratum (NAS), location services, "AI for Network" (AI4NET), and the data framework. This ensures that AI agents will not be a niche feature but an integrated part of future networks.

Pilot projects, such as the one implemented in Zhejiang (China) using the AONP (Internet-of-Agents Open Network Protocol), demonstrate the practical applicability of these concepts, including registration, authentication, authorization, and agent routing. This shows that the technologies are already ready for large-scale deployment.

Results and Prospects

Overall AI investments are growing exponentially, expected to reach $467 billion by 2030. For telecom operators who embrace AI-native core evolution, intent-driven interfaces, and agent communication frameworks, this means not only increased efficiency but also the opportunity to be at the heart of the 6G service model, meeting the expectations of customers—both consumers and enterprises, and their respective AI agents.

How to Implement This in Your Business

The shift to an AI-native architecture is not merely a technological upgrade but a strategic imperative. If your business relies on complex network infrastructures and you want to be at the forefront:

• Assess your current architecture. Determine how ready your network is to integrate AI at the core level, rather than as a simple software overlay.
• Start with pilot projects. Identify specific areas where AI agents can bring quick and measurable value, for example, in planning, quality assurance, or customer care.
• Invest in intent-driven interfaces. Move from rigid requests to models where devices and agents can signal their goals, allowing the network to proactively adapt.
• Collaborate with industry standards. Participate in 6G standard development to stay informed of the latest trends and influence future advancements.

If this case sounds like what's happening in your company, our manager can help: he'll analyze your business and niche for free and point out where an AI agent would bring a real result in your case. Message the manager