AI Is Everywhere; Value Isn’t. Telcos Should Take Note

Generative AI has moved quickly in organizations from experimentation into everyday use. It’s embedded in customer operations, supporting internal workflows and increasingly shaping how employees complete routine tasks. Usage isn’t the barrier it once was.

Superficially, this looks like progress. Employees are working faster, tasks are being completed more efficiently and AI is becoming part of day-to-day activity throughout the organization. Many business leaders would reasonably conclude that their investments are starting to deliver. But this view is incomplete.

Our research points to a more uneven reality. Some teams are enjoying clear productivity gains, but others are struggling to translate access into meaningful impact. In many cases, the employees that use AI the most encounter the greatest friction. Adoption expands, but value doesn’t scale evenly with it. Instead, productivity gains are hampered by rising friction throughout the workforce. The problem isn’t a lack of AI activity, but a lack of consistent outcomes.

Most Organizations Are Measuring AI Success Incorrectly

For many organizations, the early phase of generative AI was defined by access and experimentation. The priority was to get tools into employees’ hands, encourage usage and unlock quick productivity gains. That phase is now largely complete. Awareness is high and usage is widespread. FDM CCS Insight’s Survey: Employee Workplace Technology, 2025shows that 87% of employees use generative AI for work, and it’s now embedded in a growing number of day-to-day activities.

However, the way success is being measured hasn’t kept pace. Organizations often rely on adoption and usage as proxies for progress. If employees are using AI and reporting faster task completion, it’s assumed that value is being created. But these metrics only capture activity, not effectiveness.

High usage can mask significant variation in how AI is applied. Some employees integrate it deeply into their workflows, and others use it in more-limited or inconsistent ways. At the same time, organizational systems don’t evolve at the same pace as employee behaviour, creating a growing disconnect between what individuals can do and what the organization can support. The constraint isn’t simply how employees use AI, but how organizations are structured to support it.

The result is a false signal of progress. AI appears to be scaling because usage is increasing, but the underlying outcomes are far less consistent. This is where the concept of “workforce AI maturity”becomes important. Employees don’t progress at the same pace, and as those differences widen, so do the outcomes. Productivity gains concentrate, friction emerges and organizations find that widespread usage doesn’t translate into consistent results.

Where AI Value Starts to Break Down

These uneven outcomes are not random. As AI becomes more embedded in everyday work, patterns emerge that can be difficult to see through the lens of aggregate metrics.

Capability doesn’t develop evenly. A small group of employees captures a disproportionate share of the benefits, and others remain earlier in their adoption journey. AI may be widely available, but its impact is uneven.

At the same time, some users start to experience friction. It reaches its peak not in the early stages of adoption, where you might expect it, but once usage becomes routine. As reliance increases, workflows become more fragmented and practices begin to diverge.

As capability continues to develop, further constraints appear. More-advanced users find that organizational systems begin to limit what they can do, reflecting a growing gap between execution potential and reality.

These dynamics are easy to miss because they sit behind the curtain of high usage. From the outside, AI appears to be working. Internally, value is becoming uneven, harder to sustain and more difficult to scale.

Why This Matters for Telecom Operators

These dynamics have significant implications for how operators position themselves in the enterprise market.

Telcos are expanding beyond traditional connectivity into cloud, cybersecurity and AI-enabled services as they pursue new opportunities for growth in the enterprise sector. This reflects broader diversification strategies as outlined in FDM CCS Insight’s recent report, Operators Target New Markets in Bid to Revive Consumer Spending. At the same time, many organizations have successfully provided employees with access to AI and are now struggling to translate that access into consistent outcomes in complex environments.

This creates a potential credibility gap. As operators diversify into AI-enabled and enterprise technology services, the issues they encounter internally — such as uneven maturity, fragmented workflows and governance challenges — are likely to surface in how they support customers.

The risk isn’t simply one of execution, but of focus. Many propositions remain centred on the deployment of technology — be that tools, platforms or infrastructure — rather than on how AI is applied and scaled in practice. This creates the conditions for the same pattern: strong adoption and visible activity, but uneven and difficult-to-sustain value.

The problem has shifted. Enterprise customers don’t need broader access to AI. They need support to make it work consistently for all teams, systems and workflows. Operators that recognize this will be better positioned to differentiate; those that don’t risk addressing the wrong problem.

The Next Phase of AI

Value from AI doesn’t scale evenly. It concentrates, stalls and becomes harder to sustain as usage expands. Organizations that measure success through adoption and activity risk misreading their own progress, and the underlying issues that limit value remain unaddressed. This isn’t a marginal problem. It sits at the centre of how AI delivers impact throughout the enterprise, and organizations that misread it risk scaling activity without scaling value.

Our report, The Hidden Organizational Costs of Uneven Workforce AI Maturity, explores these dynamics in detail, including how AI maturity develops in the workforce and where value begins to break down. For organizations investing heavily in AI, understanding this gap is becoming increasingly urgent.