Hello! 👋
It’s Friday, 9th October 2026. Hello and welcome back to Bold Efforts!
I used to enjoy spending time in bookstores. Not necessarily to buy anything. Just walking around, looking at books, reading a few pages here and there. Sometimes I'd come across something interesting that I had never heard of.
I still buy books, probably more than I did back then. But now I usually know exactly what I'm buying before I even open the website.
I've been noticing this in other parts of life too.
We order groceries online, get food delivered, and have almost everything brought to our doorstep. We can watch practically any movie without leaving the couch. Even at work, we can get through an entire day of meetings without having a single conversation that wasn't scheduled in advance.
I enjoy these conveniences as much as anyone. It's hard to argue against saving time.
But I wonder what happens to all the things we used to discover by accident.
There's a word for this: serendipity. Finding something valuable without setting out to find it.
Think about the last time you had a genuinely interesting conversation with someone at work. There's a decent chance it wasn't during a formal meeting. Maybe it happened over lunch, in the corridor, or while discussing something entirely unrelated.
These interactions are difficult to justify on a calendar. You can't predict which conversation will be useful. Most probably won't be. But occasionally, one leads to an idea, a friendship, or an opportunity that wouldn't otherwise have existed.
I've been thinking about this in the context of AI.
Almost every discussion about AI eventually comes back to efficiency. How many hours can we save? How many tasks can we automate? How much faster can we get from a question to an answer?
These are important questions. I've spent plenty of time working on them myself.
But efficiency generally works best when we already know what we're trying to achieve. We have a destination, and we're looking for a better route.
The really interesting discoveries often happen before we even know what the destination is.
Many scientific discoveries, business ideas, and creative breakthroughs have emerged from unexpected observations or connections between seemingly unrelated things. Of course, they also required people who were curious enough to recognize their significance.
Perhaps that's the distinction worth remembering.
Efficiency helps us get from 1 to 10. Serendipity can help us get from 0 to 1.
Now imagine an AI assistant that becomes exceptionally good at understanding your preferences. It recommends books you're likely to enjoy, introduces you to people you're likely to agree with, and filters out information it considers irrelevant.
It might save you enormous amounts of time. It might also gradually make your world smaller.
You could become better informed about everything you already care about, while becoming less likely to encounter something completely outside those interests.
This presents an interesting challenge for anyone building AI products. Should a recommendation system always give you the most relevant result? Should an assistant always take the shortest path to completing a task? How do we preserve the possibility of discovery without simply introducing unnecessary noise?
I suspect similar questions will shape how we design offices, neighborhoods, and public spaces. Some of their value comes from interactions nobody could have planned in advance.
I don't think we need to bring back every inconvenience of the past. I certainly don't miss waiting in queues.
But perhaps, in our enthusiasm to make everything more efficient, we should be careful not to optimize away the possibility of being surprised.
And perhaps I should start visiting bookstores without a shopping list again. Thank you for reading.
Best,
Kartik
I write Bold Efforts every week to think clearly about where work and life are actually headed. If you want these essays in your inbox, you can subscribe here.