A useful conversation often leaves you with something an efficient answer cannot provide: the sense that another person has understood what matters to you. They may not have solved the problem. They may simply have noticed the part you were having trouble explaining.
That is one reason I remain optimistic about the human side of an increasingly automated world. AI can make preparation easier and routine work quicker. We can choose to use some of that capacity to become more attentive colleagues, friends and collaborators.
The choice matters. Time saved does not automatically become time well spent. It can disappear into another queue of messages just as easily as it can become a conversation that helps someone move forward.
Preparation can improve presence
Imagine using an approved tool to organise your meeting notes, identify unanswered questions or prepare a rough outline before a difficult discussion. The benefit is not that you no longer have to listen. It is that you can arrive less distracted by administration and better able to listen properly.
A relationship develops through what follows. Did you understand the concern accurately? Did you do what you said you would do? When circumstances changed, did you explain the change? Those are small tests of reliability that another person can observe over time.
In a World Economic Forum commentary, Reese Wong argues that trusted networks may become more valuable as digital output becomes easier to produce. I find the idea persuasive, while recognising it as an argument about the future rather than a settled rule for every business.
Leave room for the conversation you did not plan
Some of the most useful information arrives after the prepared questions have run out. A colleague mentions a concern they did not think was important enough to schedule. A customer explains why a technically correct answer did not help. Someone new to a field asks a question that exposes an assumption everyone else had stopped noticing.
It helps to leave a little space for those moments. Put the device aside. Ask a follow-up question. Be willing to hear something that does not fit the first version of your explanation. Attention is a practical contribution, especially when people have become accustomed to competing for it.
This is not an argument against using AI in conversations or customer service. It is an argument for understanding which parts of an interaction need a person who can take responsibility. Useful automation should make that handoff clear rather than making a human response feel like a hidden exception.
Make learning a shared activity
Relationships also give people a place to learn without having to perform certainty. A trusted colleague can say, “I do not understand this yet,” and receive an explanation rather than a judgment. That kind of environment makes adopting a new tool feel more like a shared project.
McKinsey’s research discussion on the changing workplace points out that employees’ underlying priorities have remained relatively consistent while the work changes. The human questions have not disappeared: am I learning, am I supported and can I see a worthwhile future here?
I would make space for simple forms of connection: someone experienced reviewing work with a newer colleague, a small group sharing what they have learned or a manager making a difficult decision understandable. These are habits a team can practise without a large programme or a fashionable name.
The opportunity in AI is broader than producing more material. It can help us spend less effort on avoidable friction and more effort on the people with whom we are trying to do something worthwhile. That is a future I would be glad to help build, one ordinary conversation at a time.