AI Bots in Service Organisations, Part 3: Telecommunications

From Today's Chatbots to Tomorrow's Autonomous Agents

By Sandra De Zoysa, Founder & Managing Director, CEx Consulting

This is Part 3 of a four-part series on AI bots reshaping service organisations. Part 1 laid out the framework and Part 2 looked at banking and finance. Here I turn to telecommunications, an industry I spent over two decades inside, leading customer experience for more than 20 million subscribers, and where I believe the gap between bot ambition and bot reality is widest.

Telcos were early adopters of automation, largely out of necessity: massive subscriber bases, thin margins, and support volumes no human workforce could absorb alone. That head start shows. But it has also left many telcos with a graveyard of Horizon-1 chatbots that customers actively route around.

The Bots Already Working in Telecom

Chatbots handle the highest volume of any industry here, balance and data-usage queries, bill explanations, SIM activation, plan changes. Agent labour represents up to 95% of contact centre costs, and conversational AI is projected to cut $80 billion of that industry wide. The shift worth noting is from reactive to proactive: AI now flags a likely billing dispute or network issue before the customer calls, rather than waiting for the complaint.

RPA bots are the unsung workhorses of telecom back-office, provisioning new lines, processing number portability, reconciling interconnect billing between operators. As agentic AI scales, RPA isn't disappearing; it's becoming the dependable execution layer that AI agents call on for the repeatable, rules-based steps.

Web scrapers/crawl bots support competitive intelligence and network-quality benchmarking, while social bots are heavily used for brand and sentiment monitoring, critical in a sector where a single network outage can generate a social media storm within minutes.

Trading bots have less direct relevance inside telcos themselves, though they matter for the investment arms and treasury functions of larger telecom groups managing FX exposure across markets.

The Bots on the Horizon

Autonomous agent swarms are where telecom is heading fastest. Instead of a single chatbot attempting billing, technical support and retention in one conversation, specialised agents coordinate: a triage agent, a technical-diagnostics agent that can query the network, and an escalation agent, all sharing the same customer context. Google Cloud's 2026 telecom outlook frames this as "agents for every workflow" and "agents for your customers," spanning the entire subscriber lifecycle from acquisition to churn prevention.

Empathic AI companions matter enormously here because telecom sits on some of the most emotionally charged service calls in any industry, a family losing connectivity during an emergency, a small business owner facing an unexplained bill spike. The data is consistent across industries: customers are increasingly savvy about detecting generic automation and resent it. Telcos that win will be the ones building empathy-aware handoff into the bot, not just resolution speed.

Physical service robots have a more direct near-term relevance in telecom than in banking, think automated network maintenance drones inspecting towers, or robotic assistance in high-volume retail stores and warehouses. This is a horizon worth watching for telcos with large physical infrastructure footprints.

Decentralised, blockchain-based bots are the furthest out, but telecom's role as critical infrastructure for identity and machine-to-machine transactions (IoT devices paying for their own bandwidth, for instance) makes this a more plausible near-future than in most other service industries.

What This Means for CX and Network Leaders

The questions I raise with telco leadership teams:

●      Are we still running Horizon-1 chatbots that customers have learned to distrust, and does that legacy experience colour how they perceive every newer AI investment we make?

●      Is our data foundation unified enough that a technical-support agent and a billing agent are working from the same customer truth, or are we creating new silos with new technology?

●      Where does empathy get engineered back into an increasingly automated support journey, and who owns that design decision?

Telecom has the scale and the subscriber data to lead every other service industry on agentic AI. The risk is repeating the same mistake at a bigger scale: automating for cost before automating for trust.

Next in this series - Part 4: AI Bots in Insurance.