Latest / Paths of Curiosity in an AI World / Q4 - AI in Engineering, Science and Discovery: Mysteries at Every Scale with Artemis Spyrou, PhD

Transcript

Q4 - AI in Engineering, Science and Discovery: Mysteries at Every Scale with Artemis Spyrou, PhD · Paths of Curiosity in an AI World

• 0:04
  SPEAKER_01: Welcome to the Half of the Curiosity in an AI world.
• 0:19
  SPEAKER_01: Dr.
• 0:20
  SPEAKER_01: Artemis Crew makes the short-lived nuclei from those explosions here on Earth so we can study them.
• 0:27
  SPEAKER_01: Let's get curious.
• 0:34
  SPEAKER_01: Dr.
• 0:35
  SPEAKER_01: Artemis Crew is a professor of physics at Michigan State University, and at the facility for rare isotopes, she is an experimental nuclear physicist who studies the nuclear reactions that drive stars and cellular explosions using accelerator facilities around the world.
• 0:53
  SPEAKER_01: She earned her PhD in physics from the National Technical University of Athens in 2007 and has been at Michigan State ever since.
• 1:04
  SPEAKER_01: She also has led education and outreach for the lab for several years.
• 1:09
  SPEAKER_01: And in 2021, she was elected a fellow of the American Physical Society.
• 1:15
  SPEAKER_01: Dr.
• 1:15
  SPEAKER_01: Sprue, thank you so much for joining us today.
• 1:19
  SPEAKER_00: Thank you for having me.
• 1:21
  SPEAKER_01: Your experiments depend on the machine that makes atomic nuclei that exists for only a moment.
• 1:29
  SPEAKER_01: For someone who has never heard of it, what is the facility for rare isotope beams?
• 1:34
  SPEAKER_00: Facility for rare isotope beams, or FRIP.
• 1:38
  SPEAKER_00: It's a research lab located here at Michigan State University.
• 1:44
  SPEAKER_00: And it uh it's hard at its heart, it's a large particle accelerator, and uh it produces isotopes, which are forms of elements that we cannot find naturally on Earth.
• 1:59
  SPEAKER_00: So we have to produce them in the lab in order to study them.
• 2:02
  SPEAKER_00: So those isotopes, they're rare isotopes, which means that well they're rare because you cannot find them naturally on Earth, and typically they live for short amounts of time.
• 2:14
  SPEAKER_00: It can be anything from less than a second to hundreds of years, maybe even more than that, but still we need to produce them if we want to study them.
• 2:25
  SPEAKER_00: So the reason we want to study this nuclei, there are many different reasons.
• 2:30
  SPEAKER_00: Mostly what we do here at FRIB is basic research, which means curiosity, answering questions about how the world works, about the structure of nuclei, about where the elements or how the elements are formed inside stars.
• 2:49
  SPEAKER_00: But also in the end, when we produce and we study these rare isotopes, we can also learn about their properties, which then means that we can use we, the general weed, not necessarily the scientists here at EFRI, but in general, as a society, we can find uses, and so there are uses for these rare isotopes in medicine or in art, or there are many different uh uses.
• 3:15
  SPEAKER_01: I can't wait to get into that.
• 3:18
  SPEAKER_01: But first, how would you describe what you do today to your student self?
• 3:23
  SPEAKER_00: Uh I love this question because as a student when I was uh in high school, I didn't even know that this profession existed.
• 3:33
  SPEAKER_00: Uh so I would definitely have a lot to learn uh from my current self.
• 3:40
  SPEAKER_00: Um I would say that all this information and all this knowledge that we that we uh study or students study in their textbooks, it came from somewhere that someone discovered all of these things.
• 3:58
  SPEAKER_00: And so what I do today is try to push that knowledge envelope a little bit further out.
• 4:04
  SPEAKER_00: So we go beyond what exists in uh textbooks, even what exists in scientific journals, and we go a little bit further out to discover to ext extend that knowledge.
• 4:17
  SPEAKER_01: So, what is it like to run experiments at accelerators around the world?
• 4:23
  SPEAKER_00: It's uh exciting for sure.
• 4:26
  SPEAKER_00: It's challenging uh for many reasons.
• 4:30
  SPEAKER_00: Uh in research, everything we do is new, which that's what makes it exciting.
• 4:37
  SPEAKER_00: Uh anything are in our experiments, we are the first people to see some a new piece of information.
• 4:46
  SPEAKER_00: Uh at the same time, because we are the first people to do that, we no one has done it before, so we need to figure things out.
• 4:54
  SPEAKER_00: Um and so we try to be very prepared when we run experiments.
• 4:59
  SPEAKER_00: Uh we prepare for months, sometimes years, uh plan for everything, and if this doesn't work, what then?
• 5:06
  SPEAKER_00: So we we are well prepared.
• 5:09
  SPEAKER_00: It's uh sometimes things will not work as expected, uh, and that actually sometimes it's exciting by itself because then if you if you are expecting something and you see something different, then you have to change your way of thinking uh in order to fit this new piece of information in the big picture.
• 5:30
  SPEAKER_00: In physics, this is what we try to do.
• 5:32
  SPEAKER_00: We try to explain the observations, we try to explain how nature works.
• 5:37
  SPEAKER_00: So if now there's a new observable that doesn't fit in our previous models, then we have to readjust all of the previous models so that the new piece fits.
• 5:48
  SPEAKER_00: So we do this here at Michigan State where I am, uh, but also sometimes there are other facilities around the world with complementary uh capabilities, and so we'll sometimes go to other labs to collaborate with other scientists, maybe use different equipment or maybe use different accelerators to again do our experiments and collaborate in a complementary way.
• 6:13
  SPEAKER_01: How could you walk us through your path and how you ended up studying the nuclear physics of stars?
• 6:20
  SPEAKER_00: Um as I mentioned already, when I was in high school, I didn't even know this field existed.
• 6:27
  SPEAKER_00: Uh I knew that I liked math and physics and I was good at it, and I also like to explain things, and so I was thinking maybe I'll go to college and study either math or physics and become a high school teacher.
• 6:42
  SPEAKER_00: Uh so that that's what I that was my goal going to university.
• 6:48
  SPEAKER_00: Uh I went to the University of Thessaloniki in Greece for my undergrad, and while there, first couple of years, uh we had to take every different type of physics class available, and I started to like very much the atomic physics, talking about the history of these uh discoveries in the early 1900s, the nuclear physics, being in the lab and doing measurements.
• 7:16
  SPEAKER_00: So these were some of the things that I liked very much, and I also liked astrophysics.
• 7:22
  SPEAKER_00: So then towards the end of my undergraduate career, this is when I discovered that there's a field that combines all of the things that I liked, which was the hands-on part, the nuclear physics part, and the astrophysics part, so this nuclear astrophysics field, and I thought, okay, this is where I want to be.
• 7:42
  SPEAKER_00: I also was interested in nuclear medicine, so medical applications, and I took a class when I was in my, I think my fourth year of undergrad.
• 7:51
  SPEAKER_00: I took a class and uh they took us on a field trip to a hospital to just see what the environment would look like if we would end up following that career, and I almost passed out, so I discovered that this was not quite my field.
• 8:07
  SPEAKER_00: Uh so yeah, that solidified my decision to go into astrophysics instead.
• 8:13
  SPEAKER_01: For our listeners who may not have a background in this, what makes up an atom?
• 8:19
  SPEAKER_00: So the picture that we uh we have, and it's very no one has seen an atom in a microscope, but uh based on uh experimental evidence and models and so on, the atom has a very dense central core, which is the nucleus, and then it has electrons kind of flying around the nucleus.
• 8:40
  SPEAKER_00: Uh it's a relatively large distance.
• 8:43
  SPEAKER_00: So if we would if we would put the atom to scale and say that the atom is the size of a football field, for example, then the nucleus would be just a tiny little golf ball sitting in the center of the football field.
• 8:58
  SPEAKER_00: This is how much empty space there is, this is how far away those electrons are.
• 9:02
  SPEAKER_00: But now everything is scaled down into very tiny spaces.
• 9:06
  SPEAKER_00: Uh, the nucleus itself consists of protons and neutrons.
• 9:10
  SPEAKER_00: Protons and neutrons are also not elementary particles, they also consist of smaller particles, but for the world of nuclear physics, we kind of stay there and we deal with nuclei at the level of having protons and neutrons.
• 9:23
  SPEAKER_00: So the the protons in the nucleus will define what kind of element we're talking about.
• 9:29
  SPEAKER_00: So carbon has six protons and hydrogen has one proton.
• 9:34
  SPEAKER_00: And then the number of neutrons doesn't change the element, but it changes the isotope.
• 9:40
  SPEAKER_00: So carbon, for example, with six protons and six neutrons, that makes carbon 12, and it's one of the isotopes of carbon.
• 9:48
  SPEAKER_00: But you can have seven neutrons, eight neutrons, nine neutrons, and it's still carbon.
• 9:53
  SPEAKER_00: However, its nuclear properties will change.
• 9:56
  SPEAKER_00: So carbon 12 is stable, we find it everywhere.
• 10:01
  SPEAKER_00: But if you go, for example, to carbon 14, then that lives for a few thousand years.
• 10:06
  SPEAKER_00: If you go all the way to carbon 18 or carbon 20, they still do exist, but they'll live for a fraction of a second.
• 10:14
  SPEAKER_01: So what is a rare isotope?
• 10:18
  SPEAKER_00: So rare just means that we can't find them, typically it means we can't find them naturally on Earth.
• 10:24
  SPEAKER_00: So when you get a sample from Earth, in my example with carbon, we would have find carbon 12 and carbon 13, and that's it.
• 10:34
  SPEAKER_00: Uh every every other isotope of carbon, if we want to study it uh for uh whatever application, whatever reason we want to study it for, uh we would have to make in the lab.
• 10:45
  SPEAKER_00: Uh and as I mentioned earlier, the some of these isotopes live for hundreds of years or thousands of years, and some of them live uh for a fraction of a second.
• 10:57
  SPEAKER_00: Uh however, our solar system is of the order of five billion years old, and so anything that lives for less than that, it you cannot find it naturally on Earth.
• 11:09
  SPEAKER_01: So that leads to one of the big questions.
• 11:12
  SPEAKER_01: Where do the elements and our periodic table come from?
• 11:18
  SPEAKER_00: It's uh it's a very complex and very uh it can be a very long answer for this one, so I'll keep it short.
• 11:26
  SPEAKER_00: I'm teaching actually here at Michigan State the the graduate level course of nuclear astrophysics, which goes through the whole thing in uh a full semester.
• 11:36
  SPEAKER_00: So we'll get the the two-minute version of that.
• 11:40
  SPEAKER_00: Um so after the Big Bang, what we uh find the universe, we find the universe to consist only of hydrogen and helium, a few, a little bit of lithium, and that's it.
• 11:54
  SPEAKER_00: So the the hydrogen and the helium we see uh in the universe is mostly coming from the Big Bang.
• 12:00
  SPEAKER_00: And then this is the material available for the first stars to form.
• 12:04
  SPEAKER_00: So the first stars are all formed by hydrogen and helium.
• 12:08
  SPEAKER_00: However, as a star is formed and um uh from whatever cloud is available in that part of the universe, um gravitational forces will pull the particles closer and closer together, and that then as it gets closer and closer, the central part of the star will start to warm up.
• 12:28
  SPEAKER_00: And if the temperature goes high enough, then that hydrogen that we have available in the center of the star will start to fuse.
• 12:35
  SPEAKER_00: So hydrogen fuses into helium.
• 12:38
  SPEAKER_00: Um there are many different conditions and different stages in a star's life, but during these various stages we have different nuclear reactions happening at the center of a star, and so from hydrogen fusing into helium, then we would have helium fusing and producing carbon, uh, and we kind of keep going to heavier and heavier elements during this evolution of a star.
• 13:02
  SPEAKER_00: Depending on how massive a star is, this whole process can stop earlier or can keep going.
• 13:08
  SPEAKER_00: The most massive stars will go all the way to producing iron or iron uh area nuclei in their center before this process ends.
• 13:19
  SPEAKER_00: Um, so to go and see the elements that are heavier or produce the elements heavier than iron, uh, this doesn't happen during kind of the normal evolution of a star, it happens under different conditions and different environments.
• 13:32
  SPEAKER_00: Um and um again, it's uh it's an open field of research to see to figure out exactly how these heavy elements are formed.
• 13:42
  SPEAKER_00: So this is this is what my group and many groups here at FRIP and around the world are working on to understand how the heavy elements are formed in the universe.
• 13:51
  SPEAKER_00: But we have some basic idea, they're formed mostly through neutron capture reactions, uh, either in a slow way, so it's a process called the S process, S is for slow, or in a very fast way through the R process, R is for rapid.
• 14:07
  SPEAKER_00: Um there are other processes in between, um so it is uh kind of a more complicated mechanism, and uh as I said, this is an open question.
• 14:20
  SPEAKER_01: How can an experiment on Earth teach us about an explosion in a star?
• 14:27
  SPEAKER_00: So when we want to study how stars work, uh then we need to have some basic understanding of what conditions we want to study, what kind of stellar environment, and what the conditions are in the stellar environment.
• 14:44
  SPEAKER_00: We don't necessarily know exactly what's going on, but we have a rough idea based on observations, based on our models, and based on the nuclear physics information we have already.
• 14:56
  SPEAKER_00: But based on all of this, then let's say we want to study a particular environment and there are models that say that a certain type of reaction is the important reaction happening inside that star, then we need to design an experiment that will study that reaction at conditions that will match the conditions inside the star, so namely the temperature of the star.
• 15:21
  SPEAKER_00: So what we have to do is get okay, in this stellar environment, this is the temperature that exists, and we need to translate that temperature into a velocity for our particles so that when we have that nuclear reaction happening, that it matches those same conditions.
• 15:36
  SPEAKER_00: And so this is what we do.
• 15:38
  SPEAKER_00: If you translate the temperatures in stars to lab energies, it actually comes out that it's not a very high energy.
• 15:48
  SPEAKER_00: So, for example, here at FRIB we go to energies that are 150, 200 MeV per nucleon.
• 15:57
  SPEAKER_00: These are the units that we we use.
• 16:00
  SPEAKER_00: Um, however, the stellar environment, even the the most hot environment, the supernova Nova explosion, for example, gigacelvin of energy, when you translate that into the lab, it's more like a 3 MeV per nucleon.
• 16:15
  SPEAKER_00: So it's much smaller.
• 16:17
  SPEAKER_00: However, we need we need the high energies here uh in the lab to produce their isotopes, and then we need to slow the particles down so that they are right at the astrophysical energies, and then we can measure the reactions at the right conditions that um tell us something about the stellar environment.
• 16:38
  SPEAKER_01: So briefly, how does Ephraim make the isotopes?
• 16:43
  SPEAKER_00: We always have to start with something that exists on Earth.
• 16:47
  SPEAKER_00: This is uh we can start with anything else.
• 16:50
  SPEAKER_00: Um and uh there's a particle accelerator that will get whatever it is that we start with and it will uh make it go really fast.
• 17:01
  SPEAKER_00: We are talking about 50% the speed of light.
• 17:04
  SPEAKER_00: When the particles go this fast, then they hit a target, and when they do that, the uh the dominant reaction at these energies is called fragmentation.
• 17:16
  SPEAKER_00: So your initial nucleus that is coming down the pipe, it will break into pieces, and out of all the pieces that are coming out, some of them will be stable nuclei that you can find on Earth, and we we have other ways of studying those, we don't need to do it here.
• 17:32
  SPEAKER_00: However, some of them will be these rare isotopes that we're interested in.
• 17:37
  SPEAKER_00: So then the next step is to have a very powerful filter, it's called a fragment separator, and so this filter will filter out all the things that you're not interested in and isolate the one isotope, rare isotope that we want to study in our experiment, and then send it down into the experimental station where whatever experiment we want to study is sitting there waiting.
• 18:03
  SPEAKER_00: So it can either be implanted into a detector so we can watch for its uh decay to happen, or it can heat another target nucleus so we can look for those reactions that are happening out, and then we have detectors around it to detect the reaction products or the decay products.
• 18:20
  SPEAKER_01: How do you measure a nucleus that lasts for only a moment?
• 18:26
  SPEAKER_00: It's uh it's the sensitivity of our detectors and it's the speed of these particles.
• 18:32
  SPEAKER_00: Um so when the facility produces these rare isotopes, it takes some tens of nanoseconds for the particles from production to make it into our detectors.
• 18:44
  SPEAKER_00: So if anything leaves at least some tens of nanoseconds, it will make it to our detector so we can detect it.
• 18:52
  SPEAKER_00: So this is how fast uh and then our detectors themselves are even more sensitive than the nanoseconds.
• 18:59
  SPEAKER_00: It's under nanosecond sensitivity for the detectors themselves.
• 19:03
  SPEAKER_00: And so uh a nucleus that lives for a millisecond or a second, this is a huge amount of time.
• 19:11
  SPEAKER_00: We can produce it, we can uh implant it, or have another reaction happen with it, and we have kind of the time, flexibility to do that because of the sensitivity.
• 19:22
  SPEAKER_01: How do you trust a measurement when you can only get so few atoms?
• 19:29
  SPEAKER_00: When you only measure very few, then you you better calibrate and confirm and validate your experimental setup very well.
• 19:40
  SPEAKER_00: And this is what we do.
• 19:41
  SPEAKER_00: So we will always, for an experiment, especially a new discovery experiment, where you expect maybe a handful of nuclei in a week, then at the same time, we will also measure a lot of other nuclei that we know their properties and we know what to expect about them.
• 20:00
  SPEAKER_00: They have been seen before in experiments, and so then you calibrate your your detectors and your system, and everything is you know perfectly matched, and then when you see something at a place or a location in your data where this is where you would expect the the new information to come in, then you know that you know if everything else is calibrated, and then you know this is a new discovery, this is where you see the new thing.
• 20:26
  SPEAKER_00: But yes, you have to convince yourself you can't just jump from carbon-12 that I talked about before to carbon 20 with nothing in between.
• 20:35
  SPEAKER_00: So you kind of slowly get there and you you know everything in between, you understand your system, and then you can see the new thing.
• 20:43
  SPEAKER_01: What did the 2017 Neutron Star Merger show you about where gold and other heavy elements come from?
• 20:53
  SPEAKER_00: 2017 was a was a big year.
• 20:57
  SPEAKER_00: It was an important year because of the Neutron Star Merger discovery.
• 21:03
  SPEAKER_00: Uh, just to take a step back, just to give you the picture, we talked a little bit about heavy element nucleosynthesis and how it's still an open question.
• 21:12
  SPEAKER_00: So, about half of the heavy elements we know that they are produced through this R process that I mentioned before.
• 21:19
  SPEAKER_00: However, we did not have at the time any uh confirmed astrophysical site where the R process uh could take place.
• 21:29
  SPEAKER_00: We have models and the models have make predictions and we have a good idea of what conditions to expect, but we did not have a confirmed site.
• 21:38
  SPEAKER_00: And this is a difficult thing to do because what the way that these models work is that you you have your observations, and the observations come from stars, and it's not the art process, it's an explosive process, so the whole thing is done in a couple of seconds, and then whatever was produced during this art process.
• 22:00
  SPEAKER_00: Kind of slowly decays, all the produced isotopes decay slowly, and then they end up with some stable isotopes.
• 22:08
  SPEAKER_00: And then those stable isotopes are sitting there, and billions of years later we see a signature of these isotopes in our telescopes.
• 22:16
  SPEAKER_00: So what we see today as a signature of our process is something that happened billions of years ago.
• 22:23
  SPEAKER_00: So there's a very long time lack there to get the information.
• 22:30
  SPEAKER_00: So the important thing about the 2017 event was that it was something that we could see the information coming live.
• 22:37
  SPEAKER_00: It was not live, again, it happened a long time ago, but we had the time signature, the time evolution.
• 22:43
  SPEAKER_00: So it was all of a sudden this huge amount of new information, new observations that now our models had to, we had to calibrate our models against.
• 22:52
  SPEAKER_00: It wasn't just these stellar abundances from billions of years ago.
• 22:57
  SPEAKER_00: There was a lot more information.
• 22:59
  SPEAKER_00: And so this allowed the models to be refined.
• 23:03
  SPEAKER_00: It gave us one site for the R process that is now confirmed, neutron star mergers.
• 23:09
  SPEAKER_00: Yes, we know that this is where it's happening.
• 23:12
  SPEAKER_00: It still created a lot more new questions compared to it.
• 23:20
  SPEAKER_00: So now we know neutron star mergers produce gold and other heavy elements, but we don't know if it's the only source.
• 23:29
  SPEAKER_00: Most probably now we know that it's not.
• 23:32
  SPEAKER_00: If there are other sources, how much do they contribute?
• 23:35
  SPEAKER_00: How much of these elements that are produced in the neutron star merger, how much of that is ejected out into the interstellar medium?
• 23:42
  SPEAKER_00: There are still a lot of questions unanswered in this field, but it was definitely a very exciting time, the 2017 event.
• 23:51
  SPEAKER_01: How do you estimate a nucleus you cannot measure?
• 23:55
  SPEAKER_00: You measure a lot of the signatures and a lot of the information around it.
• 24:25
  SPEAKER_00: What do they tell you about the nucleus itself?
• 24:28
  SPEAKER_00: I'll give an example.
• 24:30
  SPEAKER_00: So if let's say we have a wall and it's it's a very soft wall, so we can uh we can actually put an imprint on it.
• 24:41
  SPEAKER_00: And if you get, I don't know, a football and you shoot it at the wall, then it will give leave an imprint, and the football comes out of the wall, it doesn't stay there, but you can see that there's a hole, a football-shaped hole.
• 24:56
  SPEAKER_00: So you don't need to see the football itself to know what it looks like, how big it is, what its shape is.
• 25:03
  SPEAKER_00: You can infer that uh information from looking at the signatures, in this case, the imprint on the wall.
• 25:09
  SPEAKER_00: So it's a similar thing.
• 25:10
  SPEAKER_00: We can shoot other particles at the nuclei, we see what comes out, and then from that infer what nuclei, uh what shapes they are, what sizes they are, what they look like.
• 25:20
  SPEAKER_01: For those who are wondering how this relates to the real world, why is it so important to understand these nuclei?
• 25:29
  SPEAKER_00: There are many reasons scientists study nuclei in general and rare isotopes in particular.
• 25:36
  SPEAKER_00: And I think a big thing that drives us is curiosity.
• 25:41
  SPEAKER_00: It's it's basic this is what basic research is.
• 25:45
  SPEAKER_00: Uh it's it's the simple thing that you you're sitting and uh on your balcony in the in the evening and you look at the sky and you see the stars, and you start to wonder, hmm, why do they look like that and why do they shine and how far away are they?
• 26:03
  SPEAKER_00: And what is going on in their center?
• 26:05
  SPEAKER_00: And so these kind of questions are driving basic research, and it's the same thing with nuclei.
• 26:11
  SPEAKER_00: How big are they?
• 26:12
  SPEAKER_00: Do can we get different shapes?
• 26:14
  SPEAKER_00: What happens if I uh smash two of them together?
• 26:18
  SPEAKER_00: What's going to come out?
• 26:19
  SPEAKER_00: What happens if I smash them together, but now at a higher energy?
• 26:23
  SPEAKER_00: Um, so there are many different things, and this is what basic research is is uh answering the curiosity about how the world works.
• 26:33
  SPEAKER_00: At the same time, this is also uh many scientists are driven not by just the curiosity of how things work, but also how can I use this new information to make something useful.
• 26:45
  SPEAKER_00: Uh and so the the classic example is with with rare isotopes is that smart people saw the use of these rare isotopes in medicine, and today a lot of rare isotopes are used in different nuclear experiments, but rare isotopes specifically are used to diagnose and treat disease, like cancer, for example.
• 27:09
  SPEAKER_00: And this is to do that, you first have to really understand their properties, the properties of nuclei, so that you can use them in a safe way in whatever application.
• 27:21
  SPEAKER_01: Are machines getting better than physicists at predicting the properties of nuclei nobody has measured?
• 27:29
  SPEAKER_00: Right now I'll say no.
• 27:31
  SPEAKER_00: Uh I don't know what's going to happen in 10-20 years, but right now the machines AI is giving us good tools, I would say, but still we need to be the ones to ask the questions.
• 27:49
  SPEAKER_00: We and what questions do we want answered, and then we can use it, uh, use AI more as a tool to to help us get there.
• 27:59
  SPEAKER_00: It will help us get there faster.
• 28:01
  SPEAKER_00: Uh but uh if we as scientists are not there to ask these questions and drive the process, I don't know.
• 28:11
  SPEAKER_00: Maybe in the future it will, but right now I don't think that this is uh this is the case.
• 28:16
  SPEAKER_01: 10 to 20 years from now, what do you think someone will be doing differently in your job because of AI?
• 28:24
  SPEAKER_00: As I said, AI can be a very useful tool.
• 28:27
  SPEAKER_00: It can accelerate the the process uh a lot.
• 28:30
  SPEAKER_00: Though the one example I can give uh is programming.
• 28:35
  SPEAKER_00: It's our students when they come here to learn, undergraduate, graduate students when they come here to learn, they are training to be nuclear physicists, they're not training to be computer scientists, for example.
• 28:47
  SPEAKER_00: However, in the process to get there, they need to have the skills to program whatever it is that they need to program, from controlling our equipment to creating models and so on.
• 29:00
  SPEAKER_00: And so with AI, they can they don't need to focus on the programming so much, but they can focus on the physics.
• 29:07
  SPEAKER_00: They still we we still cannot trust that what comes out of the the AI machine will be 100% correct on the first try.
• 29:16
  SPEAKER_00: We're still the physicists, and we are still need we still need to control and diagnose and make sure that everything is correct.
• 29:24
  SPEAKER_00: However, it can make this process go much faster.
• 29:28
  SPEAKER_01: What's one concrete thing a student can do in the next year to start moving towards this kind of work?
• 29:35
  SPEAKER_00: So I think because we don't know where this is uh this is going, I would say if you asked me this question five years ago, I would definitely say you should uh go you should uh learn programming because this is something that you don't know in advance, but this is something that you will need every day in your life as a researcher, and this is the part that maybe now gets reduced.
• 30:00
  SPEAKER_00: And I think the part that is hard to get from a machine, and it really has to come from you, is the curiosity part.
• 30:07
  SPEAKER_00: So I would say try to be more curious, learn, read books, uh listen to podcasts, uh, things you know that drive your curiosity and find the things that you're interested in because you will always be better at the things that you like to spend your time on.
• 30:26
  SPEAKER_00: If you're forced to spend your time on something, then you will just do the minimum, the the least amount of work.
• 30:32
  SPEAKER_00: But if it's something that you want to do yourself, then this is uh something you'll spend more time and you'll get better and better at it.
• 30:39
  SPEAKER_01: What skill do you think matters in your job more than most students would expect?
• 30:46
  SPEAKER_00: Collaboration, I'll say.
• 30:49
  SPEAKER_00: It's um it you I think the picture we all grow up and we have the picture of the scientist isolated in their office, and then they are they just jump up and say, I found it, I discovered something.
• 31:03
  SPEAKER_00: And this is far, especially for an experimentalist, this is far from the truth.
• 31:08
  SPEAKER_00: Uh I have never in my life run an experiment by myself or even just with my own research group.
• 31:14
  SPEAKER_00: We always collaborate with different groups, and it's it in our experiments, it's not huge collaborations, it's 10-15 people collaborations.
• 31:23
  SPEAKER_00: There are other physics research where they need thousands of collaborators, but it it's I think it's almost impossible to find a case where a person will do something just by themselves, and so being able to talk to people and communicate well, this is an important skill that people don't realize is needed.
• 31:44
  SPEAKER_00: Especially these days, we are all attached to our phones and on social media and we are texting, and we are not used to so much talking face to face and try to communicate our ideas with each other, and even disagreeing with each other that's also a skill to learn that you can disagree and it's still okay as long as you you have the reasons and you can justify and you can communicate.
• 32:08
  SPEAKER_00: Um, so yeah, people communication is important.
• 32:13
  SPEAKER_01: Well, you can't run these big experiments on your own, you just can't be an expert at everything.
• 32:20
  SPEAKER_00: Yeah, correct.
• 32:22
  SPEAKER_01: So, what is the biggest question you wish in your field that someone could answer?
• 32:27
  SPEAKER_01: And is AI getting any closer to it?
• 32:32
  SPEAKER_00: So in my field of research, we're still looking at heavy element nucleosynthesis, it's a very specific question, but it's uh still very important.
• 32:39
  SPEAKER_00: It's such a basic question because half of the the elements, more than half of the elements, are in this category, these heavy elements.
• 32:47
  SPEAKER_00: Um and it's amazing to be in 2026 and we still don't have all the answers yet, considering that the field of nuclear astrophysics started in the 1950s, even before that, but in the 1950s we already had like a comprehensive theory of where all the elements are coming from.
• 33:07
  SPEAKER_00: And so, you know, you say, oh, maybe it's small changes since then, we have it all figured out, and we really don't.
• 33:15
  SPEAKER_00: Um AI, I don't think that it has any better answers than we do right now, and it will also it might help us combine information better.
• 33:28
  SPEAKER_00: Nuclear astrophysics is an interdisciplinary field, so we need the astronomical observations, many different astronomical observations.
• 33:36
  SPEAKER_00: Uh, we're in this multi-messenger era where you can get different pieces of information from astronomical observations all at once.
• 33:45
  SPEAKER_00: Um AI might help us uh analyze that data faster, and uh it might help us write or create bigger and better models for it, um, or design better bigger and better experiments, but still you still need the researchers to work together and to um to get to these answers.
• 34:07
  SPEAKER_01: And finally, I know you mentioned that curiosity is such an important skill and that you should be reading different literatures and figuring out if you like this.
• 34:18
  SPEAKER_01: Do you have any books, podcasts, or other media you would recommend for a student who wants to learn more?
• 34:24
  SPEAKER_00: Um yeah, and I don't know, maybe I can give you links to put uh um to attach to to this podcast, but again, it depends on what level there are textbooks.
• 34:36
  SPEAKER_00: Typically the textbooks are more for a graduate level, maybe a little bit advanced uh undergraduate uh uh physics level.
• 34:43
  SPEAKER_00: Um, but they're uh education material for students of all ages.
• 34:48
  SPEAKER_00: We have some at the FRIP website, for example, if someone wants to know about FRIP itself.
• 34:53
  SPEAKER_00: Um there are some videos out there that scientists recorder recorded exactly for this purpose.
• 35:00
  SPEAKER_00: Um so I I'm happy to help to send some links.
• 35:04
  SPEAKER_01: Thank you so much for joining us today.
• 35:07
  SPEAKER_01: Thank you.