Chetan Nayak
speaker
39 appearances
1 recordings
1 series
first heard Apr 2025
last heard Apr 2025
Chetan Nayak’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
It's a great question because oftentimes people think that quantum computers are just a faster version of classical computers that are sped up by orders of magnitude, as you said. But actually, they're really a very different computing paradigm. In short, what a quantum computer aims to do
is to take advantage of the underlying laws of nature, which are quantum mechanics, so that you can have what's called a qubit, replacing the basic unit of information in a class computer as a bit. It's a zero or a one. A qubit, on the other hand, like Schrodinger's cat, which could be both dead and alive at the same time, a qubit can actually be both zero and one, in a quantum superposition.
So a quantum computer takes advantage of that basic fact of nature, which although that's true of everything around us, we have the luxury of kind of forgetting about that or ignoring that as we go around our daily life. This computer screen in front of me is not both here and somewhere else, it's only here. And that's because as objects get larger, their quantum effects tend to get suppressed.
But as things get small, they actually, their quantum effects tend to get accentuated. And as Moore's Law has progressed over the last decades, the transistors on chips and the density of elements on processors has gotten so high and the transistors have gotten so small that they are getting really close to that world where quantum effects become important.
You could view that as potentially a disaster because you want your information to be a zero or a one. You don't want it to be both zero and one sometimes.
But it turns out it's also an opportunity because there are certain problems which are really difficult to solve ordinarily that a quantum computer, if we can build one of a large enough scale and stability, would be able to do relatively easily.
Well, as you said correctly, solids, liquids and gases are different states of matter. And as you continuously change, for instance, the temperature in a solid like ice, its properties change continuously. You know, you warm it up a little bit, its density changes a little bit. But then you get to the transition point.
And at that transition point, a small change in temperature leads to a huge change in its properties. And it becomes water at the melting point. And then again, at the boiling point, it becomes steam. So there are clear distinctions between the solid ice and the liquid water. There are, as it turns out, there are actually finer classifications of solids.
For instance, some solids are magnetic, some solids are non-magnetic, some solids are metallic, others are insulating, some are actually superconducting, you know, which is a remarkable phenomenon that occurs when you cool down metals. They tend to actually become better metals and better conductors as we make them colder.
But then actually there's a very special point, the critical temperature, and below that temperature, it just falls off a cliff and goes to zero. And below that temperature, the resistance is just zero. That's a superconducting state. So that's a really cool state of matter discovered early in the 20th century.
What are they? The actual device is based on an interesting combination of materials. It involves a semiconductor and a superconductor, but that alone isn't enough to get this state of matter or to make the qubits that we build on it. We cool this device down to extremely low temperatures, and that's the temperature which everything stops moving and there's no energy whatsoever in a system.
And our devices are 50 millikelvins, so 50 thousandths or 0.05 kelvin which means 0.05 degrees Celsius above absolute zero. Not the freezing point, but absolute zero, which itself is minus 273 degrees Celsius. So basically, this is really, really, really cold. And the reason is because we need to keep these qubits extremely stable, and they are more stable when they're colder.
So the fluctuations and noise that might disrupt these qubits are strongly suppressed when they're colder. The second thing that we have to do is we actually have to apply a pretty big magnetic field. So this is the kind of field where when people go into our labs, even standing outside of the fridge and outside the magnets, you have to be a little careful if you have a pacemaker, for instance.
So these are really serious, very large magnetic fields. In the units that are used here, these are two Tesla magnetic fields. So many, many thousands of times larger than the Earth's magnetic field, for instance. So these are very large magnetic fields in very low temperatures that enable us to take the raw material of the device we made and induce this new state of matter.
People have said, you know, quantum computers, useful quantum computers are decades away. We believe that they're years away. We've actually just signed a contract with DARPA, U.S. Government's Defense Advanced Research Projects Agency. I can't go into all the details, but we signed a contract to deliver something pretty serious on a pretty aggressive timeline.
And part of that process was us coming up with a very detailed plan, including not only the technical milestones, but all the other things you need to do to build something of this scale, the budgets, the hiring plans, you know, all the supply chain issues. And went over that very carefully.
I got a lot of scrutiny from DARPA and emerged at the end of that process with a plan that we think does take us to something that I think would be really cool in years and not decades. So we have a lot of confidence in our plan, and we've signed a contract, so we take that seriously and intend to deliver.
This is the longest running R&D program in Microsoft's history. So I'm coming up on my 20th anniversary with the company and I've seen this program grow from a very, you know, for the first 10 years it was a really small pilot project with just a handful of people. mostly trying to figure out if there was anything real here to quantum computing and worth the company pursuing in a big way.
We've grown over the last 10 years and especially over the last five into the kind of team that can do the science and engineering that I think is necessary to build a machine like this. Over time, we have made a number of pivots as we learn new things.
And I think as much as we like the ideas that we've developed, we certainly have the humility to realize that other people have great ideas too, and we shouldn't fall in love entirely with our own ideas. And we also learn sometimes that Something we thought was correct wasn't correct, or something that we thought would be easy turned out to be hard, and sometimes the reverse.
Showing 1–20 of 39 · page 1 of 2
Next →