Customer experience in an AI-first world has evolved from reactive issue resolution to predictive, context-aware care. Modern CX depends on understanding customer intent, sentiment, and history in real time, delivering a personalized journey that addresses friction before it escalates. While routine interactions scale effectively through conversational AI, human expertise remains vital for edge cases requiring moral reasoning, emotional intelligence, and nuanced judgment. TELUS Digital balances this through a clear automation framework: standardized, repeatable processes are automated to empower human agents with real-time assistance, ensuring operational speed does not compromise empathetic, trust-building human connections.
Indian enterprises are shifting rapidly from AI experimentation to scaled deployment, embedding generative AI into core operations to drive productivity, accelerate time-to-market, and lower defect density. Scaling AI responsibly requires a “security by design” mindset grounded in data privacy, explainable governance, and human-in-the-loop oversight to ensure system transparency and accountability. Leaders must look past viral technology trends and evaluate AI initiatives strictly by their ability to impact core business KPIs, customer satisfaction, and operational efficiency. “The organizations that will see real returns are the ones staying disciplined about linking every AI investment to actual business results,” said Archana Sharma, Senior Director, Service Delivery, TELUS Digital.
1. How is AI changing what customers expect from service delivery, and what does ‘’good CX’’ look like in an AI-first world?
Archana: In an AI-first world, ‘’good CX’’ is anticipatory, not just reactive. Customers want more than a quick resolution. They want a frictionless, personalized, journey where problems are addressed before they escalate. Speed still matters, but contextual intelligence is what sets good CX apart now. AI lets us understand a customer’s history, sentiment, and intent in real time, so every interaction feels informed rather than generic. CX has moved from transactional to relational, where the brand has a sense of what the customer needs before they’ve fully said it. Predictive care over reactive support is where good CX sits today.
2. Where does human expertise remain irreplaceable as automation scales, and how does TELUS Digital decide what to automate versus keep human-led?
Archana: TELUS Digital’s approach to AI pairs people with the right technology, equipping human agents with AI-powered tools such as real-time assistance, call summarization, and sentiment analysis so they can resolve issues faster. Routine, self-service interactions can be handled through conversational AI, which frees agents to focus on higher-value moments with customers. Human expertise still matters most when it comes to complex judgment, emotional intelligence, and high-stakes decision-making, particularly in nuanced “edge cases” where empathy makes all the difference. Our decision framework is straightforward. If a process can be standardized and scaled without compromising quality, we automate it. But if it requires moral reasoning, deep relational context, or emotional nuance, it stays human-led. This balance ensures efficiency doesn’t come at the cost of the human connection that ultimately builds trust.
3. What shifts are you seeing in how Indian enterprises approach generative AI compared to a year ago? Are companies moving past pilots into scaled deployment?
Archana: This past year has shown a clear shift from ‘AI curiosity’ to ‘AI utility’. A year ago, enterprises were largely focused on exploring what generative AI could do. Today, the focus is on outcomes. Companies are moving past pilots and are integrating GenAI deeply into core operations to help improve engineering productivity, code quality, and time-to-market. The conversation in boardrooms has evolved from ‘should we use AI’ to ‘how do we scale it for measurable business impact’. Indian enterprises have moved past experimenting at the edges and are now embedding AI into how they operate and compete.
4. What does ‘responsible AI’ look like in practice at TELUS Digital, and what should businesses prioritize as they scale AI?
Archana: Responsible AI has to be ‘security by design’, not an afterthought bolted on later. At TELUS Digital, this rests on three pillars: privacy, security, and humanity-in-the-loop. That means robust data privacy protocols, rigorous human-in-the-loop validation, and continuous upskilling of our teams to work alongside AI responsibly. For businesses scaling AI, the priority should be explainable AI and transparent governance, backed by a clear data ethics framework and real human oversight, not just a policy on paper. You need to be able to audit how a system arrived at its conclusion and confirm it operates within clear ethical and secure guardrails. Trust is earned through transparency, and that must be built into the system from day one, not retrofitted after issues arise.
5. How do you separate AI initiatives that genuinely drive business outcomes from those that are just experimentation? What should leaders watch for?
Archana: I believe it’s time we stop measuring AI usage and start measuring business value. If an AI initiative doesn’t reduce defect density, accelerate cycle times, or directly improve customer satisfaction metrics, it remains experimentation, however sophisticated it may look. My advice to leaders is to be wary of ‘shiny object syndrome’, the temptation to adopt AI tools because they are trending, rather than because they solve a specific problem. If you cannot map an AI initiative to a clear, measurable business KPI, it’s likely just noise. The organizations that will see real returns are the ones staying disciplined about linking every AI investment to actual business results.
