At the Eurasian Customer Experience Forum, I represented Flomni in a panel discussion on how AI is changing customer experience and what remains in human hands.
The discussion covered where AI already outperforms people, where customers still want a human, and which responsibilities companies are ready to hand over. We also explored disappointments with GenAI, the limits of existing service metrics, and how the work of CX teams is changing.
For me, the practical question is what happens when automation reaches its limit. An assistant can answer a routine question quickly. But when a customer has an unusual problem, repeats themselves, or needs someone to make an exception, the handoff becomes part of the product. If they have to explain everything again, the time saved on the first response means little.
This is how I think about AI at Flomni: let it handle repeatable work and help people understand the conversation before they step in. Human judgment becomes more useful when the person has the context, time, and authority to resolve the problem.
That also changes what we should measure. Response time and automation rate can look good while customers remain stuck. I care more about whether the issue was resolved, whether the customer had to come back, and how much effort the whole interaction required.