Over Half of Telecom Operators Are Already Deploying Agentic AI
According to a McKinsey survey of telecom executives conducted in December 2025, more than half of respondents are now deploying agentic AI use cases across at least one business function. That number was close to zero 18 months earlier. Meanwhile, IDC projects global AI-supporting technology spend to reach $749 billion by 2028, with two-thirds of that going directly into core operations rather than experimental pilots. For European telecom, the question is no longer whether agentic AI matters. The question is how quickly operators can move from isolated use cases to integrated, autonomous workflows.
What Agentic AI Actually Means in Telecom
Agentic AI is a class of AI systems that can observe conditions, set goals, plan multi-step actions, execute those actions across connected systems, and learn from results. This is different from generative AI, which produces content on demand but does not act autonomously.
In telecom, this distinction matters because the operational environment is distributed, real-time, and interdependent. A network fault in one region can cascade into customer complaints, truck rolls, billing disputes, and churn. Traditional AI can flag the fault. Agentic AI can detect the anomaly, diagnose the root cause, trigger a remediation workflow, notify affected customers, and update the trouble ticket — without a human initiating each step.
The transition from copilot-style AI to autonomous agent systems is what the industry now calls the shift from “AI as a helper” to “AI as a doer”.
Why European Operators Cannot Afford to Wait
European telecom faces a specific set of pressures that make agentic AI adoption structurally urgent rather than aspirational.
GSMA data from Q2 2025 shows that customer care accounts for roughly 47–50% of all telecom AI deployments globally, driven by the need to reduce cost-to-serve. Vodafone’s AI chatbot SuperTOBi already handles over 45 million interactions per month across multiple languages. These are not pilots. They are production systems operating at scale.
At the same time, telecom CAPEX is declining. IDC forecasts a 1.5% drop in global telecom CAPEX for 2026, bringing the total to $320 billion. Operators are under pressure to do more with less, and the freed-up capital is flowing toward AI infrastructure, shareholder returns, and digital plays rather than traditional network expansion.
The EU regulatory environment adds further complexity. The Gigabit Infrastructure Act, effective since November 2025, mandates accelerated fiber and 5G rollouts. Operators must meet coverage and performance obligations while simultaneously transforming their operating models. Manual processes cannot scale to meet these dual demands.
McKinsey estimates that roughly half of the telecom workforce could be restructured through collaborative human-agent automation — more than double most other industries. This is not about replacing people. It is about redirecting human effort from repetitive coordination tasks toward supervision, policy design, and exception handling.
Six Ways Agentic AI Is Being Deployed in European Telecom
1. Autonomous Customer Care Operations
Customer service represents 7–11% of a telecom operator’s total operating expenses. AI-powered digital agents are reducing call handling times by 45% and cutting service costs by 35% in documented deployments. Oliver Wyman estimates that generative AI-powered digital agents can generate over $2 trillion in value across the telecom sector, with a significant share concentrated in customer service management.
The operational model is shifting from reactive call centers to proactive, multi-channel agent systems. Telefónica’s Aura platform handles over 400 million interactions annually across 30+ channels, now augmented with generative capabilities for real-time personalization.
2. Self-Healing Network Operations
AI-powered network management reduces downtime by up to 30% and OPEX by up to 25% in documented cases. Telecom operators collectively lose around $20 billion annually to network outages and service degradations, according to TM Forum research cited by Prodapt. Agentic AI enables closed-loop automation: the system detects anomalies, diagnoses root causes, and executes remediation workflows without waiting for human intervention.
Vodafone Germany adopted semantic digital twin technology in mid-2025 to move toward autonomous network operations. Huawei’s RAN Agent, presented at MWC 2026, demonstrates intent-driven network automation using a closed-loop system covering forecasting, analysis, decision-making, and execution.
3. Predictive Infrastructure Planning
Leading operators are using AI to reprioritize capital expenditure based on real-time customer experience data rather than historical traffic models. McKinsey documents cases where operators assessed customer experience along heavily used commuting lines — affecting an estimated 5–10 million users — and reprioritized their 2025–2026 CAPEX plans accordingly.
This approach shifts CAPEX allocation from coverage-based planning to experience-based planning. The result is more targeted investment in areas where network quality directly affects retention and revenue.
4. AI-Powered Sales and Product Knowledge Agents
A growing operational gap exists in how telecom operators handle product inquiries, competitive comparisons, and pre-sales support across digital channels. Most operators still rely on static FAQ pages, siloed knowledge bases, and scripted chatbot flows that cannot adapt to complex or comparative questions.
An AI agent built on a structured knowledge base — combining internal product data, competitor intelligence, and real-time pricing — can provide accurate, persuasive, and compliant responses across website, WhatsApp, social channels, and messaging platforms. This is the approach Lab51 implements for enterprise clients. The process involves auditing and ingesting all internal product data, mapping competitor positioning from public sources, building a normalized comparison matrix, and deploying the agent across multiple customer touchpoints using a consistent knowledge layer. The outcome is a sales-ready AI agent that answers product questions with verified data, handles competitive comparisons without hallucination, and operates 24/7 across every channel where customers engage.
5. Energy Optimization and Sustainability Compliance
European operators have largely addressed Scope 1 and 2 emissions. The challenge now is Scope 3 — emissions embedded in purchased equipment and customer usage patterns. Morgan Stanley reports that AI enables telecom networks to enter “microsleep” modes during low-traffic periods, reducing energy consumption and restoring capacity as demand rises. Vodafone has reported up to 33% power reduction at 5G sites in London using AI-driven optimization.
6. Sovereign AI and Federated Infrastructure
The EURO-3C project, announced by Telefónica at MWC 2026 and backed by the European Commission, connects more than 70 organizations to build a federated cloud and AI infrastructure across borders. Agentic AI is a stated priority for this initiative. The project reflects a broader European strategy to reduce dependence on US and Chinese hyperscalers while maintaining competitive AI capabilities.
Why Acting Now Is a Strategic Requirement
Analysys Mason forecasts that less than 25% of AI tools deployed in telecom will fully deliver value in 2026. The gap between deployment and impact is where most operators stall. The differentiator is not access to AI technology. It is the ability to integrate AI into existing workflows, data architectures, and governance structures quickly and accurately.
Operators that treat agentic AI as a 2027 initiative risk compounding their technical debt. Every quarter of delay means another quarter of manual processes, rising OPEX, and lost competitive ground against peers who are already scaling autonomous operations.
The practical first step is not buying a platform. It is auditing your data, mapping your workflows, and identifying the 3–5 use cases where agentic AI can deliver measurable impact within 8–12 weeks.
The Operators That Move First Will Define the Next Era
European telecom is at an inflection point. The operators that build structured, governed, scalable agentic AI capabilities in 2026 will set the terms for competitive advantage through the end of the decade. The ones that wait will spend the next five years trying to catch up.
The technology is ready. The economics are clear. The question is execution.
