 It's hard to cheat that, but taking your warhead is fairly simple. So what does, this is another picture of what a warhead looks like. This is the delivery boss of the peace maker ICBM, which can carry 10 individually targetable warheads. So the ICBM comes out, when it comes to re-enter, it's 10 warheads that attack 10 different target simulators. Each one of them is about this big. Now the thing is that how to make sure, let's say the Russian inspectors come, how do they know that this thing is real and not empty inside? Do you open it? If you open it, you find out lots of secrets about the weapons made out of it. And the Americans would never agree to that. So verifying a warhead that's extremely difficult, because looking inside the warhead is not allowed. The countries would simply not agree to that. Airplanes is easy. Winsons is very hard. So people say that the easiest way, the only way to truly test that, to verify it, is by deflating them as they go. That's out of question. So how do we do this? How do we do verification? Without finding out national weapons secrets, without having to resort to unnecessary definitions. That's what really my work is about. So in the past, people tried to come up with different ideas as to how to verify the authenticity of a warhead before its destruction, without leaking inside. And they came up with this idea of information bearers. The idea being that you have got a warhead sitting in front of you, you bring in this system of detectors that measure neutrons and gammas coming out of the weapon itself, and then there's a computer which analyzes data and it tells you yes or no. And there's specific kind of criteria. Is there a plutonium inside, which you can sort of tell if there is a neutron inside? Yes. Is there high-explosive presence? Yes. So I should show that thing that's high-explosive. There's plutonium inside. And then keep going through that. But there's a fundamental problem with this thing. These criteria have to be extremely general. Because if they are more specific, you're going to find something about the weapon. That's not a lot. And so just to say what the paradox that you're stuck with, I put this picture of a rabbit inside a box. And let's say this box is wrapped inside a box. And you're trying to prove to someone that it is a rabbit. But you don't want them to find out the color of his fur, the length of his ears, I don't know, its level of cuteness, something like that. So how do you do that? You put some kind of information barrier. We checked and said, does it have long ears? Yes. Does it have big feet? Yes. Does it have fur retail? Yes. So it must be a rabbit. Right? It has those two things. It's got to be a rabbit, right? Right. Well, this is a picture of my dog. Okay. He has long ears. He has big feet. And he has fur retail. So the problem is that these criteria are so general, they have to be general, that essentially cannot distinguish between bunny and the rabbit. The rabbit and the dog. And that's such a good, because for these specific examples, you can take a piece of explosive, you can take a chunk of plutonium and claim it's a bomb. But the bomb is much more complex than a piece of plutonium, a piece of uranium explosion set. So how do you do this without having to resort to these electronic techniques? The other problem is that these information barriers, they have to resort to electronics and software, which you can hack, there's possible backdoors, there's various kind of things you can do. So the alternative technique that we have working on is what's called template verification methods. Template verification essentially starts in a completely different way. Rather than trying to do an absolute measurement, it tries to do a comparison. And as all physicists will tell you, comparisons are always easier and absolute measurements. Like, a symmetry measurement is much easier than a cross-section measurement. I had to do a cross-section measurement for my PhD. It was a nightmare. Then I did a symmetry measurements for my postdoc. It was much, much easier. Comparisons in general are much easier to do. You can do a lot more by simply comparing it. So the idea is that you first get an alcohol, a golden coffee, an actual warhead, which is you have authenticated to be real, that noise real. And there's some ways of doing that. For example, you, inspectors show up randomly at an ICBM site. They pick one of the warheads, and they will know that it's real, because the country needs to have real warheads on their ICBM. So if you got it off of the ICBM, it must be real. Then what you do is you take that warhead with yourself to the bunker where a bunch of warheads are sitting and now you do comparisons. Check this one against one. If you're equal to one, yes. So if this one's authenticated, this one's identical to this, then this is also authentic. So comparisons. Then once you do that, you send a dismantled one, you count them towards that country's obligations under the treaty. But there is still a big problem here. But how do you do this comparison without finding the antidote? So that's the problem. You need to come up with a comparison which has no false positives, so someone tried to fake you. And where you can essentially know information about the warhead will come up. That's a fairly difficult thing to achieve. So for that, we use essentially this technique which we very generally call physical photography. And this is kind of really the core of my talk. So for that, we use a process called nuclear resonance fluorescence. It's a process that is fairly old, but it's not very well known. You can think of what's nuclear resonance fluorescence. Think of atomic fluorescence. So come to the example of how materials at atoms can emit photons. If you take a piece of wire, you heat it up, it will emit photons due to blackboard radiation. You'll put it through a bridge, look at its spectrum, you'll see it continue, all in. If you take some kind of a gas, like a neon, and heat it up, or pass electricity through it, you'll get a very specific colors coming up, corresponding to very specific energies. And the reason is because an atom is actually a bound state, a quantum state, which can live in very specific states. In the ground state, first and second, second, second, etc. So if you excite that atom, it can transition between those states and emit photons at a very specific energy. So when looking at this energy, you can find out which gas it is. That's how, for example, we do atomic spectroscopy. If we look at the surface of the sun, we look which lines are being emitted, for there we can find out what the surface is made of. We can use this in astrophysics, we can do the material analysis, etc. It's a very powerful technique. And it gives you essentially an elemental analysis of the material, tells you which element you are looking for. Neuclearism fluorescence is a nuclear equivalent of that. So just like you had atoms, which are essentially quantum bound states, you also have nuclear, which are also quantum bound states, except that the energies are much higher. The binding energies at the nucleus is something like a million times higher than the binding energy of an atom, which is why processes of nuclear energy are so much more powerful. This is why nuclear bombs are much more powerful than explosives. So the equivalent of this thing is the equivalent of blackboard radiation is branched out, where you run an electron into material, like into metal, you end up getting a whole continuum of different energies, this is a spectrum. If you emit photons as some particular material, like uranium, you end up getting this spectrum. This is basically the probability distribution of the photons that you are going to see. This is a very clear emission of the specific light. Basically what I'm getting at is that your signal will be the fingerprint of the material you are looking for. Just think about that. You are observing something from which you can uniquely determine the material that you are looking at. And that itself is an extremely useful thing that you can use to really verify what the thing is made of. Yes. As time goes on though, it breaks down and so the signature is not going to be the same. So you have like some kind, doesn't the nuclear material itself break down? Very slowly. So but the signature won't deviate enough to where we won't be able to say this is the same exact thing. So there are problems of deviations, but it does not have to come down with the breakdown. Most weapons are produced with the same amount of, the similar amount of time, so the breakdown is fairly small. I mean it's like, you know, 10 to the minus 20 type of effects. What is really significant along the lines of what you are seeing is that the weapons are not made exactly identical to each other. So the signature will be slightly different. So you do have some slack, but it's not significant. Well, so it is still a normal question. With 90,000, there must be at least more or two that are kind of overlapping, right? So if you take two, so the thing is that how similar are the weapons with each other? If you take the pit from the uranium form of a weapon compared to uranium form of a weapon, how similar are they with each other? That's completely classified. Like it's completely secret. We don't know. There are probably very similar challenges with geometries, I suppose, make-ups, but how similar they are, that's what I don't know and that's something but that's just a little bit. So let's give this discussion as an answer for the actual question. Okay. So a few more explanations about what NRF nuclear is about is you have got some kind of a nucleus like uranium-45. It has ground state. It has a first excite state. Let's say 1733, actually it's not the first one. It's the fourth one. But if you have got a photocombing which happens to have the same energy as the transition, then you can trigger a transition to this excite state. But this state is very short-lived. Like I'm talking about picoseconds or something like that. So it will decay back down and then meet the photocombing. Your detector will count the photon of that energy and then you'll know that this is double-tilt. If it's some other material like uranium-218, which is a different isotope of uranium, well, this energy is not matched so the photon will go just right to you. And your detector will see absolutely nothing. So this is basically how you use NRF to tell the difference between materials. This is a very simple fact. There are laws of the place. So now, in reality, there's many, many, many of these states so you can have different kinds of transitions. Each one of them emitting photos of different energies which then result in this particular spectrum with, again, all these keys that this combination is unique to the isotope that you are looking at. Not just the element, the isotope. Okay, so how would you use this technique to do what we're trying to do? So the idea would be is that we'll take a brem chong beam and think of brem chong as like X-ray. It's just X-ray photons of a continuum of energies, like white light, except much higher energies. You shine that X-ray through the weapon and it goes right through. What happens is that this NRF takes place inside the weapon itself. It sort of filters the spectrum. It creates what are called absorption lines. Just like in a topic spectra that get absorption lines, get absorption lines in this NRF spectrum. And then you have a foil which is essentializes are encrypting foil which is made by the hosts. So the hosts control these. Hosts would be the country that is showing that their weapons are real before destroying them. They have a foil with materials that are specifically agreed to but be the same ones as the ones inside the weapon. Plutonium, maybe iron, maybe aluminum, maybe carbon because there's exposure to carbon or something like that. And then you're going to have secondary. You are going to have this process, NRF I described for a second time. And your detectors are going to be sitting over here so they will only see photons that are coming from the weapon itself. So they end up seeing a signature which is a convolution of what happens here and what happens over here. If you don't know what this is made out of from this signature you can never reconstruct what is made out of what is in here. This is the basic concept of physical cryptography. You get a signal which through the physical process itself is a convolution of two different materials. And because you don't know either one of them you can never uniquely identify each one of them. So one would... There's a detective, right? He's a hot-toucher. Yeah, that's true. The German detector. A German detector. It's a top-top detector. No, it's a top-top detector. They don't know what is in it. They just know what is in it. Oh yeah, a German detector. It's a top-top detector. This is a top-top detector. For example, we have a big class of people who have energy for work and they do not have energy for everything. Because they do not have the energy for whatever they do. And they have to live in a family that has energy for anything. That is called energy. If a family has energy, they do not have energy at all. They have to live a good life, they do not have energy for anything else, that is called a res containment. That is called heavy life. So there is nothing like heavy work, So how would this process take? The process that it would take is that you take your memorabilia