Telecom operators are stepping toward a future where artificial intelligence actively manages their networks, a development that could have broad implications for how mobile services are delivered and consumed. Recent demonstrations by Nokia and Amazon Web Services (AWS) have drawn attention to new technologies that promise to make network operations more responsive by giving AI the authority to control aspects of service delivery without human intervention. Industry insiders predict this technology could soon have a measurable impact on both consumer experiences and enterprise connectivity solutions, especially in regions where network resources are under pressure. Notably, businesses expecting cloud-like flexibility are watching closely, anticipating the integration of new automated approaches into telecom networks.
Earlier trials of network slicing typically involved static configurations that required significant manual input and planning, limiting real-time flexibility. Attempts to automate network management using AI have been explored in pilot projects over the last several years, but adoption progressed slowly due to regulatory uncertainty and operational risk. The latest efforts by Nokia and AWS, which incorporate agentic AI and the Amazon Bedrock platform, mark a step forward in offering more dynamic and autonomous control to telecom networks. Operators such as du and Orange reflect a continued industry push toward more adaptive, cloud-centric models, aiming to better match network supply with fluctuating demand. The evolution now includes a focus on leveraging AI not only for analytics, but for direct, instant resource allocation in live settings.
How Does the AI-Driven Slicing System Work?
The system under evaluation by Nokia, AWS, and their operator partners leverages agentic AI to monitor network metrics such as latency, throughput, congestion, and external data inputs. This allows network slices to be automatically adjusted according to situational needs, from supporting large-scale public events to prioritizing communications for emergency services. By automating what was once a manual and time-consuming process, the technology aims to improve overall service continuity and quality, benefiting both operators and end users.
What Are Operators Hoping to Achieve?
Telecom providers see potential in offering on-demand, customized connectivity stoked by smart automation, which could mean new revenue streams and greater competitiveness, especially in enterprise markets. Many large organizations have voiced expectations for network resources to behave like cloud computing, scaling up or down in real time. Orange explained,
“Our customers want flexible, strongly adaptive connectivity, similar to what they experience with cloud resources.”
Nokia also highlighted the business perspective, adding,
“Automation with AI enables us to rapidly deliver tailored network services, which is becoming increasingly vital.”
Could Cloud Providers Influence Telecom Operations Further?
Cloud companies, including AWS, are playing a growing role in telecom network modernization, as seen in the integration of Amazon Bedrock with Nokia’s slicing and automation systems. The shift to software-driven and cloud-native networks creates room for AI-based control mechanisms to be layered on top, enhancing the ability to act quickly and efficiently. However, the approach raises important questions about oversight, accountability, and regulatory compliance—concerns that operators intend to address by maintaining human involvement and running controlled pilot programs before broader deployment.
While the tested systems indicate a technological advance in network management, their practical adoption will hinge on resolving challenges around safety, reliability, and transparency. As operators refine deployments with real-world trials, industry stakeholders are watching closely to see how much autonomy AI will be entrusted with, particularly in mission-critical scenarios. The lessons learned here may set a precedent for similar deployments across various industries that rely on secure, adaptive connectivity—such as manufacturing, public safety, and entertainment.
Wider industry use of network slicing has often been constrained by slow adoption rates and concerns over complexity, but tangible progress in AI integration could accelerate momentum. Companies relying on private 5G networks, especially those with demanding performance requirements, may find increased value in consistent and scalable connectivity. For telecoms, the successful realization of these tests could encourage new business models, refreshed relationships with enterprise customers, and refined expectations of what AI can practically achieve within essential infrastructure.
