This post was published in the Common Craft Newsletter Subscribe Here
One of my favorite tech podcasts is Hard Fork. Hosts Kevin Roose of the NYT and Casey Newton often try to understand new and emerging technologies. To do so, they bring on expert guests and pepper them with questions.
Last Friday, they discussed Google's announcement of a new "quantum chip" called Willow that seemed to have amazing power. To help understand this news, they brought on Julian Kelly, Director of Hardware at Google Quantum AI. Julian and the hosts did a good job of trying to explain one of the hardest concepts in tech.
To me, the explanation problem lies in the nature of quantum mechanics, which operates at such a small scale that it appears to break the rules of what we expect. It's counter-intuitive. We don't have familiar concepts to say, "It's like that thing you learned in school." Even the scientists working on it don't claim to understand exactly what's happening, even though they can measure it precisely.
In preparation for the interview, Kevin Roose asked a chatbot to "explain quantum computing to an idiot". What he got back was an interesting analogy that uses a comparison to a normal computer:
This state of being both heads and tails at once is called superposition and it's one of the fundamental ideas.
Kelly points out that a quantum computer can solve new, different kinds of problems because it's a fundamentally different kind of computing:
Part of the problem with quantum computing is that the qubits are fragile. Google's new chip is an improvement because it makes the qubits less fragile, allowing them to scale more easily. This is called "error correction"
Once again, the hosts asked AI how to think about this idea:
Kelly added another analogy that relates to the threshold that quantum computing has to reach in order to improve versus becoming more chaotic:
It's important to know that quantum computing is a field of research that will likely take years to produce products or solutions for everyday life.
The Lesson
Yes, quantum computing is exciting and confusing. But the lesson I learned from this interview is that most ideas can be learned by asking an AI chatbot to explain a subject "to an idiot." I do this all the time, and it's incredibly helpful. I say this as someone whose career is based on explanations.