This piece is about self-reflection and the power of questioning assumptions. It's for me first and before anyone else. I hope this helps. And if other people have a better formula for applying this kind of thinking, I hope they share it.
How often do we treat an inherited way of doing something as though it were the only way?
I know everyone is tired of Elon stories. But one story captures the way I would love to think. Bear with me. I promise a good end.
I first read Elon Musk’s biography by Ashlee Vance around ten years ago, before the hype surrounding him reached its current level.
A version of this goes into the book: After speaking with aerospace manufacturers around the world, Musk found that buying a rocket could cost as much as $65 million. He considered the price absurd. Instead of accepting it as an unavoidable fact of the aerospace industry, he broke the problem down into its most basic parts.
“I tend to approach things from a physics framework. Physics teaches you to reason from first principles rather than by analogy. So I asked, what is a rocket made of? Aerospace-grade aluminum alloys, some titanium, copper, and carbon fiber. Then I asked, what is the value of those materials on the commodity market? It turned out the raw materials cost about two percent of the typical price of a rocket.”
The lesson I get from this is not that a rocket should cost only as much as its raw materials. Engineering, testing, labour, coordination, and risk are all real costs. The lesson is that the current price of something—and the current way of producing it—should not automatically be mistaken for a fundamental constraint.
That is the difference between reasoning by analogy and reasoning from first principles. Reasoning by analogy begins with what already exists. It asks how we can make the current product cheaper, faster, or slightly better. First-principles thinking begins with the outcome we want. It asks what must be true, which constraints are genuinely unavoidable, and which ones exist only because of earlier decisions.
This does not guarantee innovation. Nor does it mean that every convention is wrong. Many conventions exist for good reasons. But it forces us to examine those reasons instead of inheriting them without question.
This way of thinking feels particularly relevant now that useful intelligence is becoming cheap and available on demand.
Many companies are responding to AI by adding it to what they already have. They add a chatbot, a writing assistant, or another button to an existing workflow. In many cases, that will be helpful. But it still begins with the assumption that the existing product and workflow should remain largely unchanged.
A first-principles approach starts somewhere else: what is the user actually trying to accomplish, and which parts of the current process are still necessary?
Perhaps the answer is not another chatbot. It might be removing forms, eliminating hand-offs, generating the first draft automatically, or completing an entire task that previously required several people and several days. The opportunity may be to redesign the workflow, not simply decorate it with AI.
The same question can be applied beyond technology. If I were designing my career today, would I follow the same path? If I were creating my daily routine from scratch, would I organise it this way? If I were building this company now, with everything that is currently possible, which assumptions would I keep?
Perhaps the most useful question after AI is not, “How can I use this technology in what I already do?”
It is this: “If I were starting today, would I still do it this way?”