How often do we talk about AI in a way that actually helps the people trying to use it? That question has been on my mind for a while, so when Olga Lagunova invited me onto her Practically Tomorrow podcast, it felt like the perfect moment to flip the script. Instead of treating AI like a spectacle, we looked at what happens when you focus on the parts that deliver value quietly in the background.
Olga is someone I have always admired for her approach to technology. In her role as Chief Product and Technology Officer at GoTo, she has spent years thinking about how AI can support small and midsize companies that do not have the luxury of large data science teams or experimental budgets. Her outlook cuts through the noise, and our conversation followed the same pattern. We explored how AI matters only when it becomes simple to use, predictable in its outcomes, and woven into everyday tasks without fanfare. It is the opposite of the big model announcements we see every week, yet it is far closer to the reality that most firms live in.
We also spent time on a topic I know many leaders are wrestling with: how agentic AI will fit into everyday operations. Olga talked through this shift with clarity, describing how these systems can quietly take on repetitive or time-consuming tasks while still giving teams complete oversight. I found her perspective refreshing because it connects ambition with responsibility in a way that feels achievable for companies that are still early in their AI journey. There is a growing temptation for teams to adopt tools informally, often called shadow AI, which carries its own risks. Our discussion examined how leaders can sustain experimentation without losing sight of governance and trust.
Another part of the conversation that stood out for me was the ongoing debate between building AI internally and buying proven tools. It is a dilemma that every organisation faces, and Olga offered a balanced approach to cost, capability, and long-term maintenance. For many teams, the most brilliant move will be choosing ready-made solutions that deliver value quickly, especially when resources are limited. Throughout the episode, we kept coming back to a simple idea: AI projects succeed only when they are tied to real problems and measured by real outcomes, rather than vague ambition.
Being a guest on Olga’s show was a reminder that we need more conversations like this. AI is shaping the direction of work at a pace that is hard to absorb, yet most people are still trying to work out what is practical today. If you want a grounded take on where the field is heading and how to make it useful in a business setting right now, our episode of Practically Tomorrow is a good place to start.
Let me know what you think, and how your own organisation is approaching this next wave of AI adoption.