Insights

The reality of AI expectations in telcos

Artificial intelligence (AI) has become one of the most powerful—and polarizing—topics in the telecommunications industry. From boardrooms to investor calls, AI expectations in telecom are often framed around rapid monetization, breakthrough customer experiences, and near-term revenue growth. Yet as enthusiasm accelerates, many organizations are discovering a widening gap between ambition and execution.

The reality is that AI monetization in telcos does not unfold all at once. Instead, it follows a deliberate sequence—one that begins deep within network operations, long before customer-facing use cases can deliver sustainable value. Understanding this progression is critical for telecom leaders seeking to turn AI investment into measurable outcomes.

DIVIDER

AI monetization begins with network cost reduction

Across the industry, AI is frequently positioned as a growth engine. Personalized offers, intelligent recommendations, and next-best-action strategies dominate many conversations about AI value. While these initiatives have clear strategic merit, they are rarely where telcos realize their first—and most reliable—returns. In many cases, expectations for rapid revenue uplift overlook where AI can have the most immediate and measurable impact.

In practice, AI monetization in telcos starts with cost reduction in telecom networks. Modern networks are shaped by unprecedented network complexity, driven by 5G evolution, virtualization, cloud-native architectures, and distributed edge environments. This complexity has dramatically increased the cost of operating networks, making manual processes inefficient, slow to scale, and increasingly error prone.

This is where network operations automation delivers immediate value. AI-driven predictive analytics in telecom can anticipate failures before they disrupt service, reducing unplanned outages and costly reactive maintenance. Intelligent fault correlation minimizes manual intervention by rapidly identifying root causes across distributed systems. Automated network optimization improves capacity utilization and energy efficiency, helping operators do more with existing assets. Together, these capabilities directly improve operational efficiency while reducing operating expenses.

Crucially, these early wins do more than improve the cost base—they build organizational trust in AI. By demonstrating that AI can safely and consistently influence operational outcomes, telcos gain the confidence needed to expand AI’s role beyond optimization and into more visible, customer-facing domains.

DIVIDER

Why must cost reduction come before churn reduction

Reducing customer churn remains one of the most visible priorities for telecom leaders, and AI is often positioned as the primary lever. Predictive churn analytics, personalized retention offers, and AI-driven engagement platforms promise to identify at-risk customers and intervene proactively. These capabilities are valuable—but only when built on the right operational foundation.

Churn reduction in telecom is rarely driven by engagement alone. It is fundamentally shaped by service reliability, performance consistency, and how quickly issues are resolved. Dropped calls, inconsistent data speeds, recurring outages, and slow fault resolution erode customer trust long before a customer receives a personalized message or offer.

AI can help identify which customers are likely to churn, but it cannot compensate for systemic operational instability. In some cases, premature personalization can even amplify dissatisfaction by highlighting problems customers already experience but feel have not been addressed.

This is why cost reduction through network optimization typically precedes meaningful churn reduction. When AI is first applied to stabilize and optimize operations, it addresses the root causes of dissatisfaction rather than just the symptoms. A more predictable network, faster resolution times, and fewer service disruptions create the conditions under which AI-driven customer experience initiatives can deliver credible, lasting impact.

DIVIDER

Operational signal quality is the hidden dependency of CX AI

One of the least discussed—but most critical—factors in AI success is operational signal quality. AI models depend on accurate, timely, and consistent data to make reliable decisions. In telecom, that data originates across highly distributed and dynamic systems.

Customer experience AI relies on signals from network telemetry, transport performance, service platforms, and customer interactions. When these signals are delayed, inconsistent, or distorted, AI models struggle to produce meaningful insights. Predictions become less accurate, recommendations lose relevance, and automated actions introduce risk.

This challenge is particularly acute in environments where latency and jitter vary unpredictably. In such cases, even advanced AI-driven network operations operate on partial or outdated views of reality. Improving operational signal quality requires more than data aggregation; it depends on predictable network behavior, consistent telemetry delivery, and validated performance. Without these foundations, CX AI remains aspirational rather than operationally impactful.

DIVIDER

Why AI expectations often outpace execution readiness

The gap between AI expectations in telecom and realized value is rarely due to a lack of innovation. Telcos are experimenting aggressively with machine learning, generative AI, and automation platforms. The challenge lies in execution readiness.

AI systems assume predictable behavior between observation and action. Yet many telecom networks were built on best-effort principles that introduce variability and uncertainty. As a result, organizations hesitate to allow AI to control mission-critical functions, limiting autonomy and slowing progress.

This hesitation is understandable—but it also explains why AI adoption in telecom often plateaus after early pilots. Without the right operational foundations, ambition outpaces reality.

DIVIDER

A sequenced path to sustainable AI value

Successful telcos recognize that telecom digital transformation through AI is not a single leap, but a progression—one that builds value incrementally while reducing operational and organizational risk. While the end goal may be autonomous networks and differentiated digital services, reaching that destination requires a deliberate sequence that aligns AI ambition with execution readiness.

The first phase focuses on AI in telecom networks and operations, where the need—and the opportunity—is greatest. This is where AI delivers its fastest and most measurable returns. Network environments are data-rich, cost-intensive, and extraordinarily complex, making them ideal candidates for automation and optimization. AI-driven network optimization improves capacity utilization, reduces energy consumption, and stabilizes performance. Predictive maintenance and intelligent fault management reduce downtime and manual intervention. Together, these capabilities improve operational efficiency, enhance predictability, and generate tangible cost savings.

Critically, this phase does more than reduce expenses—it establishes trust. By allowing AI to influence operational decisions in controlled, high-impact areas, telcos gain confidence in both the technology and the governance models required to manage it. This trust is essential before AI can be extended into more visible and risk-sensitive domains.

The second phase builds on these operational gains to strengthen customer-facing initiatives. With a more stable and predictable network, the quality of operational signals improves. Telemetry becomes more consistent, performance anomalies are easier to detect, and root causes are resolved faster. In this environment, AI-driven customer experience initiatives become far more effective.

Churn reduction strategies, for example, benefit from both improved data quality and genuine service improvements. Predictive models become more accurate because they reflect real customer conditions, not noise introduced by operational instability. Interventions feel credible because they are supported by meaningful improvements in service reliability, not just personalized messaging. At this stage, AI helps translate operational excellence into customer loyalty.

The third phase is where AI evolves from an efficiency tool into a growth platform. Once AI is trusted across operations and customer experience, telcos can confidently deploy it to support autonomous networks, launch new digital services, and enable differentiated experiences at scale. AI becomes embedded across the organization, supporting real-time decision-making, dynamic service creation, and ecosystem-based innovation that extends beyond traditional connectivity.

Importantly, each phase reinforces the next. Cost reduction enables reinvestment. Stability enables innovation. Trust enables autonomy. Skipping steps in this sequence rarely accelerates outcomes. More often, it introduces fragility. Organizations that attempt to monetize AI through customer-facing innovation before stabilizing operations encounter inconsistent results, stalled pilots, and skepticism from both customers and internal teams. Complexity increases faster than capability, eroding confidence in AI programs and slowing long-term progress.

A sequenced approach does not slow transformation, it sustains it. By aligning AI initiatives with organizational readiness and operational maturity, telcos can move faster with less risk, ensuring that AI delivers enduring value rather than short-lived experimentation.

DIVIDER

A pragmatic perspective for telecom leaders

AI is neither a silver bullet nor a short-term experiment. It is a foundational capability that reshapes how networks are built, operated, and monetized. Realizing its value requires realistic expectations and disciplined execution.

For telecom leaders, the key questions are clear:

• Are we applying AI where it delivers the fastest and most reliable value?
• Are our operational foundations strong enough to support CX automation?
• Are we sequencing AI monetization in telcos to build trust and momentum?

At UST, we help CSPs align AI adoption in telecom with operational reality—starting with network optimization and progressing toward customer-centric innovation. The reality of AI expectations is not about lowering ambition; it is about unlocking value in the right order.

For telecom leaders ready to move from AI experimentation to measurable impact, the next step is clear: assess your operational readiness, strengthen the foundations, and build a roadmap that turns AI potential into sustainable performance with UST.