Originality and creativity needs friction

One of the most memorable moments from my master’s thesis research happened during my very first interview.

I asked an interviewee whether he had any general concerns about AI and how it might affect his work. His answer was not about job losses, inaccurate information, or the technology becoming too powerful. Instead, he raised a quieter concern: that we might become too homogeneous in our thinking, gradually losing the unique perspectives that make our ideas valuable. I’ve found myself returning to that thought.

AI gives us unprecedented access to information, suggestions, and new ways of approaching problems. It can help us overcome a blank page, explore unfamiliar topics, and generate ideas in seconds. But if we increasingly rely on AI to shape our thinking, could we also be narrowing the range of ideas we consider?

AI makes it easy to get started. Faced with a difficult problem, we can ask for suggestions and receive a structured list of plausible solutions almost instantly. Instead of exploring a problem from different angles, we may start refining an idea AI has already supplied. Rather than questioning the assumptions behind a suggestion, we may accept its framing of the problem. Our thinking becomes more efficient, but potentially less exploratory.

This does not mean AI always produces conventional ideas or that everyone using it will reach the same conclusions. But when we repeatedly use AI to guide our starting points, we risk giving its suggestions disproportionate influence over what we consider possible.

Some of our most interesting ideas emerge from uncertainty, frustration, and unexpected connections. We spend time exploring a problem, follow an unusual line of thought, or arrive at a solution that initially seems counterintuitive. AI can support this process, but it can also make it tempting to skip the uncertainty altogether.