AI is now integrated at the heart of network assurance and customer operations, tracking performance in real-time and launching automated actions across domains. As telecoms operators integrate fibre, mobile RAN and an increasingly large satellite component, planning is becoming less about a fixed blueprint and more about real-time adaptation, requiring near-real-time decision-making. That’s a radical departure from the past, writes Anna Ribeiro.
Telecoms network planning used to be paced by procurement cycles and regulatory approvals. Capacity projections were based on past traffic patterns, assessed quarterly and translated into a multi-year capital plan. Long lead times and fixed assumptions defined the model, though that no longer holds now.
With AI tracking networks and satellite capacity activated on demand, planning cycles are compressing from months to hours. This is not an incremental improvement, but more of a structural reset of how networks are run.
AI becomes operational muscle, not lab experiment
Operators are scaling AI across network assurance and customer operations. According to Analysys Mason, telecoms operators are accelerating AI investments across infrastructure, automation and customer engagement functions, moving from pilots to scaled deployments that underpin core operations.
Omdia’s research shows real-time analytics and autonomous network operations rising fast. 41% of communications service providers now see agentic AI as a key driver of autonomous network management, including diagnostics, optimisation and fault resolution at scale. This marks a clear shift from early AI experiments toward systems that can detect and act on performance issues across service assurance workflows.
Meanwhile, Gartner forecasts that agentic AI will autonomously resolve up to 80% of common customer service issues within the next few years, fundamentally altering the economics of call centres.
That is important to know for planning. When customer complaints, churn signals and quality degradations are detected and triaged at scale, the feedback loop between the network and the customer collapses. Rather than wait for quarterly net promoter score (NPS) reviews, operators can identify congestion in near real-time, automatically simulate rerouting scenarios, initiate capacity moves through radio, fibre or satellite layers and close the incident loop without escalating to humans. Obviously, planning is a continuous, not an episodic thing.
Satellite shifts from backup to programmable fabric
Hybrid connectivity is reshaping topology decisions. Satellite is no longer a resilience insurance solution. Kaleido Intelligence forecasts the number of direct-to-device satellite IoT connections to reach tens of millions by 2030, which shows rapid adoption of satellite technology into overall connectivity strategies, while Juniper Research forecasts significant growth in satellite IoT connections in the coming years as the geographical area covered decreases and costs decline.
The satellite connectivity market is rapidly moving from a niche resilience solution to a hybrid network solution. According to the Omdia report ‘Satellite IoT Market Landscape – 2025,’ the global satellite IoT connections market is expected to register a 23.8% CAGR from 2023 to 2030, as standards evolve, costs decline and enterprises seek reliable coverage where terrestrial networks fall short. This expansion underscores how operators must plan beyond traditional terrestrial resources, integrating satellites as programmable capacity that AI orchestration engines can activate on demand, bridging coverage gaps and reducing planning latency.
The research also notes that while only around 10% of global IoT enterprises had initially planned to use satellite, falling cost and integration barriers now position satellite as a scalable complement to cellular and fibre networks, enabling fast-moving operators to exploit with real-time planning and analytics.
With satellite becoming commercially viable at scale, orchestration platforms need to treat the terrestrial and non-terrestrial networks as one pool of capacity. AI-enabled orchestration gives operators the ability to dynamically route traffic between terrestrial backhaul and satellite connections, assign bandwidth according to live demand signals, prioritise enterprise SLAs during peak network traffic and deploy satellite capacity in reaction to weather or infrastructure disturbance. This convergence compresses traditional planning timelines. Capacity that once required months of negotiation can now be triggered algorithmically.
Collapse of planning cycles
AI-driven satellite orchestration is no longer experimental but a core operational lever reshaping how hybrid networks are managed and governed, with the shift now visible in boardroom metrics and budget priorities.
Analysys Mason’s research shows strong telco interest in AI for operational transformation, with automation no longer confined to pilots but expanding across network domains. Omdia expects investment in cloud-native infrastructure and software, the foundation for AI and automation, to grow from US$17.4 billion in 2025 to US$24.8 billion by 2030 globally, highlighting growth pace in programmable, AI-ready investment environments.
Operators can now run continuous digital twin simulations and automated scenario modelling over terrestrial as well as satellite links, compressing planning cycles into near real-time orchestration. Heavy Reading/Omdia data shows that an increasing proportion of networks can today realise autonomous assurance operations, with edge and cloud integration further accelerating decision cycles. This shift is a fundamental reset in how modern telcos design and manage hybrid connectivity.
With continuous planning, closed-loop simulations now run in real-time across terrestrial and satellite networks, while AI-driven orchestration dynamically reallocates capacity with minimal human input. Agentic AI also feeds customer experience data directly into automated assurance workflows, tightening the response cycle. This leads to compressed planning horizons, with digital twin simulations, automated triggers and near real-time service level agreement (SLA) monitoring replacing months-long upgrade cycles and annual budgeting rhythms.
Satellite and AI reset telecoms governance
AI-driven planning is revealing structural inconsistency in traditional telecoms operating models. Real-time orchestration pushes powers toward the edge and regulatory and capital supervision sit at the core, out of place with speed and oversight. As networks are becoming software, product teams have more control over how much capacity and how many service options they offer, without long engineering cycles.
There also needs to be a change in how we price. Simon-Kucher emphasises the increasing adoption of value- and outcome-based pricing in AI markets, while EY highlights importance of governance models that enable agility without compromising telecoms accountability.
Assurance, meanwhile, is evolving from reactive to predictive. Frost & Sullivan and ABI Research report on the growing penetration of AI-based analytics and predictive maintenance, while Transforma Insights brings attention to automation’s significance in handling the increasing complexity of multi-domain networks. AI is not a layer on top of telecom; AI is the telecoms control layer.
Strategic inflection point
Agile planning in the age of AI-driven satellite orchestration is not simply about faster networks. It is about faster judgment. Operators that succeed will embed AI deeply into assurance and customer operations, treat satellite and terrestrial infrastructure as a unified capacity pool, redesign governance to keep pace with machine-speed decision cycle and align pricing models with dynamic service delivery.
The pivot is awkward as it challenges decades of norms for hierarchical planning and sequential decision making. But with the growth of hybrid connectivity landscapes and increasing customer expectations, near real-time planning has become a necessity, not an option. It is rapidly becoming operating baseline for the AI-native telco.
Evidently, agile planning in a hybrid connectivity world is not a luxury; but more of a survival skill. Telco operators are able to see the entire stack and run tests on options rather than wait for results. That’s less waste and a better customer experience, but it also means telcos need to rethink governance, pricing and talent models. Winners will be those who combine rapid, automated orchestration with clear human accountability and a commercial model that captures these new sources of value.
Anna Ribeiro