We are living through an explosion of information.
Social media, search engines, recommendation algorithms, advertisements, and now AI agents are constantly deciding what information to put in front of us. Much of it is personalized. Much of it is designed to be relevant—or at least relevant enough to keep our attention.
This raises a question that I think is becoming increasingly important:
What information is actually relevant to us, given our current situation?
The Information We Want vs. the Information We Are Given
There is an interesting tension between two kinds of information.
The first is information we want to know. It usually comes from our current situation: a problem we are trying to solve, a decision we need to make, something we want to understand, or a direction in which we want to grow.
The second is information that is pushed toward us.
YouTube recommends another video. LinkedIn presents another post. An advertisement introduces a product. A platform predicts what will keep us engaged. Increasingly, AI systems can personalize this process even further.
I don't necessarily think this is bad. In many cases, it is simply a smart way for a platform or business to present information and create value. Recommendation systems can also introduce us to ideas we would never have searched for ourselves.
But we should recognize that these two types of information have different origins and motivations.
One begins with our intention.
The other often begins with someone else's intention.
Understanding that difference matters.
Don't Just Filter Information. Channel It.
The common response to information overload is to build better filters.
But I think channeling may be a more useful concept than filtering.
Filtering asks:
Should I keep this information or throw it away?
Channeling asks:
Where should this information go, and how might it become useful?
Not everything needs our immediate attention. Some information is highly relevant to the problem in front of us. Some may be useful later. Some gives us broader context. And some teaches us something indirectly.
For example, I might see an advertisement that I personally dislike. I could simply dismiss it as irrelevant. But I could also ask why I reacted negatively to it.
Was it too aggressive? Did I feel manipulated? What marketing strategy was being used? Would I want to use that strategy for my own product?
Suddenly, information I didn't ask for has become useful—not because I accepted its intended message, but because I channeled it differently.
That distinction is important.
We don't necessarily need to treat unwanted information as garbage. We need to decide what role, if any, it should have in our thinking.
Relevance Is Dynamic
There is another complication: our channels themselves cannot remain fixed.
The categories through which we understand the world need to evolve.
AI is already making the boundaries between many professions and disciplines less clear. Problems that once belonged neatly to one field increasingly require knowledge from several fields.
This means learning cannot simply be:
Receive information → put it into an existing category.
Sometimes new information should make us question the categories themselves.
Perhaps two subjects we previously considered unrelated should now be connected. Perhaps an old mental model is no longer useful. Perhaps a problem requires us to reorganize what we know around a completely different question.
In this sense, learning is not only about acquiring more information.
Learning is also about continuously redesigning the channels through which information becomes meaningful.
AI Makes This Even More Important
AI can help us tremendously with this process.
We can customize AI systems according to our preferences, goals, requirements, and circumstances. An AI agent could help us search, summarize, organize, prioritize, and connect information at a scale that would be difficult to manage ourselves.
But there is a deeper issue.
The AI we customize is partly a reflection of how we already think information should be channeled.
If our questions are narrow, our AI may help us become more efficiently narrow.
If our assumptions are wrong, personalization may repeatedly reinforce those assumptions.
If we optimize only for what appears immediately relevant, we may lose the unexpected information that produces inspiration.
So the challenge is not simply to build a better AI filter.
It is to build a better relationship between human intention, information, and AI.
AI can help us organize information, but we still need to examine why we consider something relevant in the first place.
Learning Begins With Better Questions
When information arrives, perhaps we should develop the habit of asking:
- Did I choose to receive this information, or was it chosen for me?
- What is the intention behind presenting it to me?
- Is it relevant to a problem I am trying to solve right now?
- If it is not immediately relevant, is there another way it could be useful?
- Does this information fit my existing way of thinking—or should it make me reconsider that structure?
And perhaps most importantly:
What do I actually want to know right now, and why?
The more powerful our information systems become, the more important these questions become.
The future of learning may not be about consuming more information or constructing increasingly aggressive filters against it.
It may be about becoming better at channeling information with intention—while remaining willing to redesign those channels as our problems, environments, and understanding change.
AI can help us do that.
But ultimately, we still have to decide what deserves our attention, how it connects to what we already know, and what we are going to do with it.
In an age of abundant information, knowing more is easy. Knowing what matters—and why—may become one of the most important skills we can learn.
So let's ask ourselves: what information are we seeking because of our intention, and what information is reaching us because of someone else’s intention? Regardless of where it came from, what should we do with it?