transcript

Speaker 1:
[00:03] You're listening to The Skeptics' Guide to the Universe, your escape to reality.

Speaker 2:
[00:09] Hello and welcome to The Skeptics' Guide to the Universe. Today is Thursday, April 16th, 2026 and this is your host, Steven Novella. Joining me this week are Bob Novella.

Speaker 3:
[00:18] Hey everybody.

Speaker 2:
[00:19] Karen Santamaria.

Speaker 4:
[00:20] Howdy.

Speaker 2:
[00:21] Jay Novella.

Speaker 3:
[00:21] Hey guys.

Speaker 2:
[00:22] And Evan Bernstein.

Speaker 5:
[00:23] Good afternoon, everyone.

Speaker 2:
[00:25] Evan, you finally emerged from your tax help.

Speaker 5:
[00:29] I liken it to a cocoon, you know, when a bug or something. Yeah, chrysalis. Thank you. An insect has to go into this state and makes a hard shell around them. And only whatever exists in their world is inside of that shell. And then crunch, crunch on April 15th. At the end of the day, you can finally start to break out.

Speaker 2:
[00:46] Now you're saying you're a butterfly. Is that what I'm hearing?

Speaker 5:
[00:48] Absolutely. I'm spreading my wings.

Speaker 4:
[00:51] Or a moth.

Speaker 2:
[00:53] If you were a moth, what kind of moth would you be?

Speaker 5:
[00:56] Not the Death's Head moth.

Speaker 4:
[00:58] Oh, I would.

Speaker 5:
[00:58] I'd be the Death's Head moth.

Speaker 3:
[00:59] I'd be Mothra.

Speaker 2:
[01:00] Mothra?

Speaker 6:
[01:01] Nice joke.

Speaker 4:
[01:02] I'm over here Googling prettiest moths.

Speaker 2:
[01:04] Okay, it's easy. Lunar moth.

Speaker 4:
[01:07] Oh yeah, those are the big furry ones, right?

Speaker 6:
[01:08] They're big boys.

Speaker 5:
[01:09] Oh, those, is that the one that Gandalf had in Fellowship of the Ring?

Speaker 2:
[01:15] No.

Speaker 5:
[01:15] Remember he talked to a moth and sent it away?

Speaker 4:
[01:17] What about the ones that look like they have birds on the tips of their...

Speaker 6:
[01:21] They got big eyes?

Speaker 4:
[01:22] There's a lot like that, actually. There's a lot of moth.

Speaker 5:
[01:24] The kind of moths you don't want are the ones that, you know, hide in your closet. Pantry moths, are they called?

Speaker 3:
[01:29] Yeah, pantry moths are horrible if you get them.

Speaker 4:
[01:32] Oh yeah, don't you have to like burn down your house?

Speaker 2:
[01:35] No, you haven't had them once. It was basically like a one-year campaign of search and destroy. Oh my god. And you also have to put all food that comes into your house or that already is in your house in the freezer.

Speaker 4:
[01:49] Oh my god, it's like having bed bugs.

Speaker 5:
[01:52] To stop the eggs from hatching?

Speaker 2:
[01:54] That will kill them, that will freeze and kill them. So things like a box of pasta, forget about it. There's no way you can keep them out of there. But if you put it in the freezer for three days, you know you're not bringing it into your house. And if they were in there, they're dead. So you got to do that. Basically everything's got to pass through the freezer. And then you also have to just, you know, just every time you think you might see one, you kill it. And you have to look in the corners of your pantries and look for their little egg sacs and you got to eradicate them constantly until eventually you'll just stop seeing them.

Speaker 4:
[02:26] That is awful.

Speaker 5:
[02:27] Oh, and I thought tax season was uncomfortable. Man, I'm gonna deal with pantry moths.

Speaker 4:
[02:31] Oh, the Atlas Moth, that's the one.

Speaker 5:
[02:33] The Atlas Moth, I remember that.

Speaker 4:
[02:34] Atlas Moth, that's my pick.

Speaker 2:
[02:36] The Madagascan Luna, Moon Moth is very pretty. That's no moth. Wow, that's my new favorite. So I got the Luna Moth and the Madagascan Moon Moth. Those are my two favorites.

Speaker 4:
[02:47] Oh yeah, that one's cool too. Check out, it's also known as the Comet Moth. Check out the Atlas, Steve, I think you'd like it too. Cause the top wings are very mimic, mimic-y. They look like snakes almost.

Speaker 3:
[02:59] I'm kind of grossed out by moths. I mean, they're not in the area of favorite insect for sure.

Speaker 4:
[03:04] Really? Oh, I love them. They're so like, I don't know, witchy.

Speaker 2:
[03:08] They're fuzzy butterflies.

Speaker 4:
[03:10] Yeah.

Speaker 5:
[03:11] Well, you know, we have a day to celebrate moths in the United States. It's the, what, the second Sunday in May, that Mothers Day. Oh, no. I was waiting.

Speaker 4:
[03:19] I was like, where is he going with this?

Speaker 2:
[03:22] That was too bad to anticipate, right?

Speaker 5:
[03:25] Thank you.

Speaker 3:
[03:26] I did not see that coming.

Speaker 5:
[03:27] Oh, high compliments, guys. I appreciate that.

Speaker 4:
[03:31] We're going to chalk that up to tax brain. He's free now. He's free.

Speaker 5:
[03:35] I am free. I am free. This was a tough season. There were lots of changes to the tax codes that happened mid-year in 2025. So the IRS wasn't quite ready to handle everything. And in fact, we had a couple of cases, some unusual cases, that they wouldn't accept the returns from some of our clients because of these very specific things having to do with those tax laws. So then these clients, we have to go back to them and tell them, we have to file manually on paper as opposed to electronically. What? Yeah. Yeah. In a couple of cases, we had to do that. We had more rejects this year of electronic filings than we've had in many, in many years. And we attributed a lot to that. Plus, the problem with the IRS in the United States right now is it's woefully understaffed. You know, there's been cuts everywhere, everywhere. And it has all these downstream impacts on everything having to do with IRS.

Speaker 4:
[04:33] So I assumed that it would have the opposite effect. Like, you guys remember, I don't know if any of you have had to re-up your global entry in the last several years.

Speaker 5:
[04:42] I'm supposed to. I haven't yet.

Speaker 4:
[04:44] When I was due to re-up my global entry, it was like, you need to make a new appointment. And then it would show me the calendar and it was booked for all of the dates. Like, it just didn't have any open.

Speaker 3:
[04:53] It just was booked.

Speaker 4:
[04:54] Yeah. And I was like, well, how do I do this then? And then out of the blue, I just got a new card in the mail and they were like, congrats, we renewed you for another 10 years. I figured the IRS worked kind of the same way. They were like, yeah, just like pass it through. We don't have time to check, but.

Speaker 5:
[05:08] Well, normally, and in prior years, they have done things like that in which they pretty much let everything go through and they just sorted out on the other side. But this year was just a real challenging year.

Speaker 4:
[05:20] Yikes.

Speaker 5:
[05:20] The states actually did a better job than IRS did.

Speaker 4:
[05:23] Okay. Can I ask you a personal, like, you know, I'm a psychologist, Steve's a neurologist. Like, you know, often we joke, do as I say, not as I do. Do you file your taxes early or are you, like, doing it on the last day also?

Speaker 5:
[05:40] Mine was the third to last tax return I did. I did it yesterday. Now, I had already pre-done a lot of it, right? So I had gamed out exactly where I was going to be. I had all my documents ready to go. So I was not in a position where I was scrambling to find information. It was really just entering the data and hitting the button.

Speaker 4:
[05:59] That's hilarious.

Speaker 5:
[06:00] But yeah, no, I just could not find a free moment. The last four weeks, I tell you, I averaged because Jennifer, my wife and I went over. Last night, we did a recap of how my schedule was the last four weeks. I was not home much. I was working anywhere. I was out of my house anywhere from 16 to 18 hours a day every day.

Speaker 3:
[06:20] Oh damn.

Speaker 6:
[06:20] That's crazy.

Speaker 5:
[06:21] The fact that I was able to come back once a week and participate in SGU was actually quite remarkable. I don't know.

Speaker 6:
[06:27] I missed out.

Speaker 4:
[06:28] We were your break. That's saying something.

Speaker 5:
[06:31] So, but you know, hey, this is my passion. This is my love. I love doing this. So I don't want to miss it for anything. So I'm very motivated.

Speaker 2:
[06:40] Oh, you mean the SGU, not taxes.

Speaker 3:
[06:42] Yes. I was too confused there for a second, too.

Speaker 4:
[06:45] Deep in there about taxes being your passion, but yeah, you're right.

Speaker 5:
[06:48] Yeah. You know, it's fine. I can do the job. I would not describe it as my passion. Certainly it's not my identity.

Speaker 2:
[06:55] Well, welcome back to the living, Evan. Thank you.

Speaker 5:
[06:57] Happy to be here.

Speaker 2:
[06:59] Jay, last week we were on the phone together, watching the Artemis 2 mission splash down in the Pacific. So this is you're completing your series, live covering sort of the Artemis 2 mission. So give us the recap. What happened? How did it all go?

Speaker 3:
[07:17] Yeah. So it went very well. We definitely were watching. I was calling Bob and Steve as things were taking off and coming back to Earth. But it was a very, very successful mission. The bottom line is this was a test mission. It absolutely wasn't trying to accomplish anything other than getting data to find out how well the entire stack performs. At the top level, the launch, the trans lunar flight, the lunar flyby, the reentry, the parachute descent, splash down and recovery, they all succeeded, which is great. And we've been doing, we've done enough of these where we have a ton of background on all of that, but this was largely a new system. There are definitely going to be details for each one of these where they could make improvements and tweaks and everything, but there was no major problems with any of that. The Orion capsule itself, that's where people are. That appeared to have functioned really well. It completely did its job of shuttling people through deep space. So we have a whole set of data now coming that wasn't on Artemis 1 because there were people on board. NASA says the crew tested life support, cabin atmosphere monitoring, food preparation, waste systems, communications, navigations, stowage, suit operations, emergency procedures. They also performed manual piloting demonstrations, and these are very important because the future missions, particularly when we land on the moon, they're going to involve very complicated spacecraft operations, maneuvering, docking, all of these different things. So they had to test that those systems were working properly. And from the testing that they did, things are looking good. But again, NASA is going to be digging into this for a very long time. So there's going to be more details coming out that we don't have now. So in general, everything worked well. And Orion did its basic job that it was created to do. There were a few things that are worth talking about that happened. One of them is this. NASA, we have a toilet problem. The toilet partially failed. So what happened? There are two different things that they had in the toilet. There's a urine collection, and then there was the solid waste collection.

Speaker 5:
[09:34] And the tidy bowl man.

Speaker 3:
[09:35] Yeah, the urine side didn't work. It was down. And what the crew ended up having to do was use CCUs. These are collapsible contingency urinals. Basically, they're plastic bags. They pee into those. Then they squish that liquid into a system that the Orion has where it can push liquids into outer space. There was a fan problem. They fixed it during the flight.

Speaker 2:
[10:01] Jay, did this shit hit the fan?

Speaker 3:
[10:03] The shit did not hit the fan. It didn't. The pee also did not hit the fan. They did fix it. NASA figured out and the crew figured out how to get that system working again, so it was a temporary problem, but again, I'm sure because it broke that there are things that NASA is going to do to make sure that doesn't happen again. NASA also planned but ended up skipping a demo showing how the crew could build a temporary radiation shelter inside of Orion in case of a solar storm or something unexpected. So it's like a pop-up tent type of situation. They quickly put this thing together, they get inside of it and it gives them much more protection from anything going on that has to do with solar radiation.

Speaker 5:
[10:42] It's interesting, I wonder why they just can't build that into...

Speaker 2:
[10:45] Because it takes up space.

Speaker 6:
[10:47] I wonder, Jay, did they do what they did in Apollo 13? Like, okay, we have this important problem. Here's all the things that they have at their disposal. Let's figure out how to fix this. And then when they fix it on the ground, they tell them, all right, this is what you do. I wonder if it was even at that level of... I mean, it wasn't a life-or-death thing.

Speaker 3:
[11:03] For the shielding?

Speaker 6:
[11:04] No, for the crapper.

Speaker 3:
[11:05] I mean, I'm sure that they have a list of every single thing that's on there, and the engineers on the ground would be in a similar situation if there was an emergency. But I felt like it was much more of a, like, open this up and move this thing to the left type of fix. You know, it didn't seem to be a major problem. But I'm sure they're always capable of doing some heavy lifting when it comes to being creative and figuring out, you know, solutions to problems that they didn't already figure out how to fix. But that wasn't the case here. Before launch, you guys might have noticed that the rocket had a helium flow problem, which they fixed before the launch, of course, took place. This was simply a displaced seal between the ground-to-rocket quick disconnect. It's basically a pipe that feeds the rocket helium and the flow wasn't happening and then engineers figured it out and they fixed it quickly. But of course, NASA is also now turning that into like, we have to figure out why that happened and you know, they're 100% error correction here with NASA. Now, the heat shield, the heat shield is a big guy here. This is a big player. It was talked about a lot. I talked about it a lot. The heat shield is what keeps the Orion capsule, which is again, this is where the humans are. They're inside this capsule. When they re-enter the Earth's atmosphere, the heat shield is an ablative material that burns away as it protects the capsule from heat. And on Artemis 1, there was some char loss, which is essentially, like, if you just think of it as them painting on, imagine if it's a layer of black protective paint. It's not that, but I'm just using it as something to think about. And that protective layer is thick enough that as it enters the Earth's atmosphere, it wears away. And there's still enough of it there to protect, but it is meant to displace heat by burning away as they go. Artemis 1 had some issues. There was some of it that was flaking off. It wasn't working properly. They made an adjustment to the re-entry. On the surface, as far as the general public knows, it worked very well. It did work very well. But we don't know the details, and NASA is going to take a lot of time to analyze that to see how well it performed and if they do need to make a final decision to fix it or upgrade it or whatever.

Speaker 6:
[13:13] Now you know, if you remember, the shuttle had the shuttle tiles, and they weren't ablative, I don't think. So I wonder why they went in a different direction.

Speaker 3:
[13:23] I don't know if there were or not, Bob. I think they were though, because NASA had to replace tiles. I don't know how extensively the replacement was, but I know that they did replace tiles. So I guess there was some wear and tear on them, but I don't know if it was ablative.

Speaker 2:
[13:37] No, they were not. They were technically not ablative. But they would crack and break down or whatever. They all didn't survive every reentry.

Speaker 3:
[13:47] I do remember seeing or reading something where there was a person whose job it was to refashion those tiles for the space shuttle and literally had to handcraft them to fit perfectly. It wasn't done by a machine.

Speaker 5:
[13:59] Each tile has a specific dimension.

Speaker 3:
[14:02] So here's the questions about the heat shield, the ablative material that need to be answered. Did it behave exactly as predicted? Did it erode more than expected? How much margin was there? Meaning how much of it was left at the end of that reentry? Was there enough to make it safe? Did the reentry profile perform as they hoped? It seems like it did. And did it reveal any concerning items for Artemis 3? Which is essentially a summary of everything I just listed. But bottom line is we'll find out all that. But on first blush, it worked. NASA also has a list of things that they, data they collected about the astronauts themselves. So they wanted to, they tracked and wanted to get more information on astronauts' sleep, their stress, their cognition, their teamwork, their immune response, their general physiology and the amount of radiation that they were exposed to. And then what effect would or could that have on them? So overall, double thumbs up. Everything worked. It worked well enough to make the mission a complete success. They did, in the end, they did test an enormous amount of things and they have an incredible amount of collected data that they have. We didn't lose the Artemis 2 capsule, right? Orion made it back and it made it back in very good condition. So they're going to get data on everything. And we've talked about this. People were saying how low res the images were when they were sending them back to Earth. Yeah, of course, they're sending lower res images from the capsule because it's very difficult to send data over, a lot of data over a very long distance. So they were sending back data that will get here and that can be processed in a quick enough way where we can see it before the Orion capsule comes back. But now they have access to the full res pictures of everything, all video. And awesome, right? So yeah, we're going to see all that. That's going to be coming out.

Speaker 2:
[15:55] Yeah, just search on like best pictures from Artemis too. There's some stunning, stunning images.

Speaker 5:
[16:00] Forget it. Those are images of the century so far, some of these shots.

Speaker 2:
[16:04] That's my new wallpaper now. Image of the Earth from the far side of the moon.

Speaker 5:
[16:09] The Earthrise.

Speaker 2:
[16:09] And then I threw in a picture of Artemis there. That's not a real picture.

Speaker 6:
[16:12] Yeah. Real quick, guys, I just saw something quick about the difference between the shuttle tiles and the ablative Artemis. Yes, it's single use. It's absolutely single use, as we said, and designed to burn away. But the reason why they went from a tile to this avcodes or is it AV code, this avcodes ablative material, is that the lunar re-entry is higher energy. So these shuttle tiles are meant for low earth orbit. They're not meant for lunar re-entry. And so more heat is generated. So you need something apparently ablative and better than the tiles.

Speaker 3:
[16:47] Yeah.

Speaker 2:
[16:48] Yeah. I also have more details on the solar radiation fort that they make. So what they basically do is just make a pillow fort. They stack up all of their supplies and equipment and food and water into a little cocoon. And they go in there and they stay there for as long as they need to.

Speaker 5:
[17:10] I know what that feels like.

Speaker 6:
[17:13] Why do they do this, Steve? What's the context here?

Speaker 5:
[17:15] We're in the case of a solar...

Speaker 2:
[17:16] There's going to be increased solar radiation. That's their extra radiation protection is to basically just build a fort out of all their stuff and get inside of it.

Speaker 5:
[17:25] So there's stuff, the material that make, that comprises that stuff has properties to it that help. That's interesting.

Speaker 6:
[17:33] Wait, they don't have a shielded section of the craft?

Speaker 2:
[17:36] Well, the craft is shielded. Bob, a shielded section, there's one section. This is a little capsule, right? There's no section.

Speaker 6:
[17:44] Oh, so you're talking about the landing aspect of the...

Speaker 2:
[17:46] The Orion capsule.

Speaker 6:
[17:47] Okay.

Speaker 2:
[17:48] If at any point during the journey, there's like, there's going to be a solar... They're going to be hit by a CME or a solar flare or something. That's too much radiation. That'll exceed their safe limits. So the capsule itself is shielded, but not shielded well enough for that kind of intense radiation. So then they stack up all their stuff, get inside of it for the duration, for extra protection from a radiation event. Evan, to answer your question, water is a great radiation shield. And so anything with a high water content would be good. Plus anything, I guess, that's like metal or whatever would also be adding to protection. It's just basically to reduce the amount of radiation as much as possible.

Speaker 5:
[18:30] Yeah, it makes sense.

Speaker 2:
[18:31] But that's what it can't be, because it can't be up all the time because they're just sitting inside a little fort, you know. All right, Kara, tell us how artificial intelligence is affecting biology research.

Speaker 4:
[18:42] Yeah, so we've talked about this a few times in the past, but I feel like this is one of those topics that's often maybe ancillary to whatever the main story that we're covering is. So I thought that I would dig a little bit deeper into an article that was published recently in The Conversation by Dr. Stephen D. Turner, who is an associate professor of data science at the University of Virginia. As he says, I'm a data scientist who studies genomics and biosecurity, and I research how AI is reshaping biological research and what safeguards that demands. So I thought we would talk about the two sides of the coin here, because very often on the SGU, when we discuss technology, we are tasked with really digging deep into what all of the incredible benefits could be of this technology, but also every time a new technology allows us to do some pretty incredible things, it also has a potential dark side, right? Like what happens if this gets into the wrong hands? What happens if this is used by nefarious actors? So that's really what this article that he wrote is all about. So he mentions how basically it all starts with a conversation about open AI, along with a biotech company called Ginkgo Bioworks, who just this year announced that GPT-5 had run on its own, autonomously, not just run, but actually designed and run 36,000 biological experiments. So how does a large language model run a biological experiment? Because I think often we kind of gloss over the specifics. And when you hear the word biological experiment, you think like wetware, right? Like biology. And that is what they're talking about. And I think that's an important point. Because sometimes when we talk about studies on this show, what we're really talking about are studies where things are modeled or simulated, right? That would be a perfectly cromulent description of what an LLM could do, is they could simulate a bunch of different biological experiments. In this case, they actually built a robotic laboratory that was all accessible through the cloud. Does that make sense? And there's pictures. And I remember actually way back when I was in graduate school the first time. So that would have been 2007 is when I got my masters in the neurobiolab. We had a robot that looked somewhat similar. We were doing a lot of electrophysics experiments. There was a lot of chemistry involved. And we had a little robotic arm that could pipette, that could lift up tubes, move them to other places, put them on the belly dancer to mix them, put them in centrifuges and put little pH test strips and things like that in there. This is very similar. You see a robotic arm, but here it's controlled remotely. And the difference is it's not just people putting in the inputs. The large language model is actually learning and developing the experimentation as it goes. So what kinds of experiments do you think that this would be really helpful for? We've mentioned it a million times on the show.

Speaker 2:
[22:08] Drug development.

Speaker 4:
[22:09] Yeah. Whether it's developing specific exogenous compounds to bind to receptors in the body or it's actually developing new proteins, for example, or just working towards that intersection between what is already in the body, what is endogenous to the body, what is outside of the body that we can make, like drugs. Exogenous? Yeah. Exogenous. Exactly. Yeah. Like drugs. But also, I think even the word drug at this point, Steve, would you agree, is sort of like, we need almost like a bigger word than the word drug because now we're talking about all sorts of like synthetic biological compounds and different elements and smart drugs and even potentially biological weapons that all sort of act like drugs in a way.

Speaker 2:
[23:06] Yeah. I mean, the drug has different definitions based on context. Drug from a regulatory point of view in the United States is anything that treats or cures a disease. Right.

Speaker 4:
[23:17] That's therapeutic. Yeah.

Speaker 2:
[23:18] Yeah. If you'd mention a disease, it's by definition a drug. It gets regulated as a drug, but it doesn't necessarily mean a chemical pharmaceutical. That's only one category. As you say, there are protein therapeutics. There's lots of different things, antibodies, monoclonal antibodies, etc. So yeah, it's sort of the bigger biological therapeutics, which has many subcategories.

Speaker 5:
[23:41] We need something catchier than that.

Speaker 4:
[23:43] Yeah, the biological.

Speaker 5:
[23:44] Biophares or something.

Speaker 4:
[23:46] Well, then let's look at the flip side of that because the way that biological therapeutics work and there's myriad ways that they can work depending on, like you mentioned, if we're talking about monoclonal antibodies or we're talking about synthesizing new proteins, blah, blah, blah. But I think the key word that you mentioned there and that's what makes them fit within that drug or be related to that drug category is therapeutic. But of course, any technology, any protocol that we use to make something that could help, could also potentially harm, right? And not just because of side effects or because it wasn't tested in full, but intentionally developing compounds that when they bind to certain receptors in the body or when they cause downstream changes to the genetic coding will intentionally cause harm. So now we're talking about bio weapons, right? So what this data scientist talks about basically is that for all of our biomedical history, we had bench work. Anybody who's ever worked in a biolab has done classic traditional bench work and they've had to learn how to use all the machines. They've had to learn how to test the pH and the osmolarity and pull out all the different buffers and compounds and mix together things and blah, blah, blah, blah, blah. This is part of the experience of working in a lab. Part of it is designing your experiment, coming up with all the protocols, figuring out how you're going to measure everything, and then how you're going to iterate as you go. The other part is the dexterity component. It's actually being good at doing the physical lab work. Often you have lab techs who are involved in that, a lot of graduate students and undergraduate students. But that part is more and more, we are capable of completely taking that part over with machines. That doesn't feel risky to most people, right? It actually feels like a boon. So people introduce error, but a robot can do it faster, more efficiently, with less waste, with less contamination. That all sounds great. But then what happens when the design part of the experiment starts to fall into the hands of AI. Now, we're applying all of those same improvements, faster, more efficient, cleaner, with less waste, to a protocol that used to take hours to days to months to years, now can happen in seconds to minutes. And so what does that mean for biosecurity? And this is a real concern. So, you know, they call this the dual use problem. You know, the same AI tools that could cause great benefit to an entire people can also be misused and cause real harm. They can be repurposed, but they can also be, you know, very intentionally used in that way. So we've got a few studies already that have been, you know, published and are available about ways that this has been done. So we're not just talking, you know, pontificating and saying, what could happen, but a few studies where they're like, okay, let's see if this actually can and has happened. So I've got a study here. I can cite where protein evolution was enhanced using large language models and they were able to improve enzyme activity, mutation rate, and evolution significantly from traditional methods. We've got a study here where viral spread was optimized, so that's fun, right? So the AI was able to figure out how to make the virus as good at spreading through a population as possible. Yeah, that's good.

Speaker 2:
[27:54] That's gain of function research, which is illegal.

Speaker 4:
[27:56] Yeah, yeah. And we've got also, how can we recover live viruses from synthetic DNA, there are large language models that can just walk somebody through that. And so the main point here, there are a lot of main points, but one of the main points that stuck out to me, it was salient for me, is there was a study that showed basically what LLMs offer, even beyond ease of use, more efficiency, improved timelines for expert researchers, is that individuals without expertise are more readily able to use these tools. And that's a real concern, because before there was a certain kind of bottleneck to doing bioterrorism, that meant you just kind of had to be smart enough to do it, right? Like you had to at least have the training to be able to use these tools efficiently and effectively, but now you don't even need that. So here is a statement by the Center for Security and Emerging Technology that was put out at three years ago talking about bio risk. They said that without AI, a scientifically knowledgeable user can follow existing protocols and produce known existing pathogens and toxins. They can also develop novel or modified pathogens and toxins through directed experimentation. With AI, a scientifically knowledgeable user can use biological design tools to design new or modified pathogens and toxins or to evade screening or enhance production. This could increase design efficiency, reduce physical burden, and then chat bots could help brainstorm new approaches. So this is 2023. And then they mention that a scientifically naive user without AI often is just doing beginner-friendly, you know, anarchist cookbook style stuff. And they're probably more likely to produce a known existing pathogen or toxin. They're not likely to modify something new or do too much cutting edge work. And with AI, through the use of these chat bots, they're probably still more likely to produce a known pathogen or toxin, but they may lower the perceived barrier. So more actors may engage in bio risk. Now, this has been blown wide open with LLMs entering the picture. Because now a naive user, and there are studies that are beginning to compile to show this, a naive user can produce novel and, let's just say, dangerous on a lot of levels, different pathogens. And specifically, there's a lot of research now into the development of proteins. Proteins before were a huge bottleneck because they're complicated. Proteins are one of the most complex macromolecules in nature, and they do so much. Proteins are just, they're ubiquitous. They're involved in so much metabolic activity.

Speaker 6:
[31:05] Everything.

Speaker 4:
[31:06] So much.

Speaker 6:
[31:06] They're all in. If somebody says to you, guess what this tissue is in the body? If you say protein, you're probably going to be right.

Speaker 4:
[31:12] Yeah, exactly. Most every enzyme is a protein. Proteins do a lot of stuff. So I did a little bit more digging, and I found a really great kind of position paper from an individual ethicist who writes about technology and the risks of powerful AI. He basically detailed some of the benefits and some of the concerns, and some of the things that we should be thinking about and highlighting. So these are the things he said, when we think about the benefits, it's important to discuss risk in a careful and well-considered manner. I think it's critical to number one, avoid doomerism, not just in the sense of believing doom is inevitable, but also more generally thinking about AI risks in a more religious way. He said we should acknowledge uncertainty, of course we need to do this, we should intervene as surgically as possible. So sometimes broad restrictions and regulations actually can do more harm than good because they can prevent legitimate research from taking place. But then he said what should we actually be worried about? Well, number one, autonomy risks. What are the intentions and goals of the country or the actor who is utilizing this technology? Is it hostile? What values does it share? Could it militarily dominate the world? Blah, blah, blah. Misuse for destruction. Assume that they follow instructions. Could existing rogue actors who want to cause destruction, like user manipulate an existing armed force in that country to do this? Is there a misuse for seizing power? How could this affect economic disruption? Then of course, as we often talk about, what are the indirect effects of utilizing this new technology? There was a Biden era executive order in 2023 on the safe, secure, and trustworthy development and use of artificial intelligence, which had a lot of potential regulations and safety nets put in place. But what do you guys think happened to that?

Speaker 2:
[33:31] We went bye-bye?

Speaker 4:
[33:32] Yeah. Well, the Trump administration just fully revoked it, so that's fun. We do see that a lot of individual AI companies are putting out policy statements, are trying to introduce their own safety checks. But of course, that's self-regulation, and we know that that's not going to ultimately be the standard that we need. Anthropic has activated its highest safety tier when it released its most advanced model in 2025. OpenAI updated its preparedness framework. So we do see that people are working on this and they are concerned about this. But for example, we have a biological weapons convention, an international treaty that prohibits production and use of bioweapons. Well, it was agreed upon and published in 1975. The word AI is nowhere in it.

Speaker 3:
[34:24] Right.

Speaker 4:
[34:25] You know? And so we have to be really careful about the fact that some governing rules, they don't account for the speed at which this can happen, and they absolutely don't address it specifically in their language. So yeah, cheaper, more accessible, more naive users can potentially cause a lot more harm, a lot more quickly, and our regulation and our policy is just not keeping up.

Speaker 3:
[34:53] Well, yeah.

Speaker 2:
[34:54] So Kara, it's funny because we both chose an AI news item for this week. You could pretty much take everything you just said and swap out my topic, and it's the exact same discussion.

Speaker 4:
[35:05] That's amazing. Well, then let's dive in. So we can just talk.

Speaker 2:
[35:09] I think it's interesting to compare these two topics. So but first, let me ask you guys a question. Do any of you know what the term vibe coding refers to?

Speaker 4:
[35:17] Yes.

Speaker 5:
[35:18] Vibes, code, anyone but Jay?

Speaker 6:
[35:20] Code or coding?

Speaker 5:
[35:21] C-O-D-I-N-G.

Speaker 4:
[35:22] So is that just like kind of like writing code based on like how you feel about it?

Speaker 2:
[35:27] Nope. Oh, okay.

Speaker 4:
[35:28] Code with the vibe?

Speaker 3:
[35:31] You know, it's software, it's a software development approach where, you know, it's basically like using natural language, right, to code.

Speaker 4:
[35:37] Oh, so it's like, I don't need, I don't have to be a coder. I can just go, hey, hey.

Speaker 5:
[35:42] You can use English.

Speaker 6:
[35:42] That's what AI does when you talk to it, when it makes code out of your words.

Speaker 5:
[35:46] Oh, that would be like, best thing.

Speaker 2:
[35:47] It's non-coders coding using AI by just telling the AI, make me an app that does this. And then you just natural language, describe exactly what you want and then AI makes it for you.

Speaker 4:
[35:59] But let's be honest, that's freaking amazing.

Speaker 2:
[36:01] Of course it's amazing.

Speaker 4:
[36:02] It's also like the best use for AI because that's like what it does.

Speaker 2:
[36:05] All right, but what do you think the downsides might be of that?

Speaker 3:
[36:09] The developers don't really understand the inner workings.

Speaker 2:
[36:12] Yeah, right. So you have naïve people who don't have the skills. They didn't. So again, partly the coding universe is controlled by the fact that relatively few highly trained people are doing it. But now if we open up coding to everybody, what could be the unintended consequences of that? And as Jay and Bob were saying, these are now, not that it isn't awesome that anybody can create some code for themselves, but what do you think the quality control is of that code? Zero.

Speaker 4:
[36:43] Of course, we don't know where the weaknesses are.

Speaker 2:
[36:45] Yeah. What's the security vulnerabilities of that code? How is that going to integrate with all the other code that's out there, et cetera, et cetera? It's the same kind of problem opens up. So this news item actually focuses around a company called Anthropic.

Speaker 5:
[36:59] Have you guys heard anything about that in the news lately?

Speaker 2:
[37:02] This has been in the news partly because they're fighting with the Pentagon. And this is an interesting overlap, I think, between that episode and what's going on now. So the Pentagon, Pete Hegseth, was mad at Anthropic because Anthropic said, you know, there's a couple of guardrails built into the software that the Pentagon's already using from Anthropic, the Claude software. And there are two red lines or guardrails. And the Anthropic CEO, Dario Amodi, said, we're absolutely not compromising on these. One is you can't use our software for massive domestic surveillance. And he said, this is due to, this is just an ethical issue. We just ethically do not think that our, this, we don't, we don't want our software being used for this purpose. And the second one was we do not want our software being used to target weapons without a human in the loop, for military targeting by itself, you know, completely 100% AI military targeting with no human who was in the loop and able to review it, right? And they said this is, this is not so much of an ethical issue as our software is not reliable enough for this. It still can hallucinate. And, you know, so the reliability is not high enough to not have a human in the loop. So the software doesn't do either of those two things. Now, Pete Hexeth, you know, and the Pentagon's, he thought that this was just the company being quote unquote, woke, which is a ridiculous thing to say. And, but their real position is that, well, we're only using the software for legal purposes. We should be allowed to use it for any legal purpose. And we decide what it gets used for, not the CEO of some company, you know, gets to decide what it's used for. That's it, you know, it sort of brings up a very interesting discussion about where this power should be. And of course, you know, my feelings about it are affected by the current people who are, would be making these decisions on the federal government side. And also, but not that I'm like happy to leave, to be at the tender mercies of tech bros either. You know what I mean? It's like, I'm not really happy with either answer, but for right now, I think the more conservative answer of sure, leave the guardrails in place. Yeah, why wouldn't you? Anyway, that's being fought about in the court, and that's not really what I'm talking about, but the Anthropic is now at the center of another, maybe even bigger issue. And that is another piece of software that they're coming out with called Mythos. And Mythos is coding software. And they claim they're about 12 to 18 months ahead of the competition. And this is the most powerful coding software ever made, right?

Speaker 6:
[39:54] This is AI.

Speaker 2:
[39:56] Oh my god. This is AI powered coding that is ahead of anything else, according to Anthropic. But here's the concern.

Speaker 6:
[40:06] Wait, give me some details. What does that mean? How are they ahead? In what way?

Speaker 2:
[40:09] It's just better. It's more powerful. It's just needs more accurate, everything. But here's the thing. When they were testing it, they discovered that the ability to write code also gave it another ability that's ahead of everybody else.

Speaker 3:
[40:24] It can read minds?

Speaker 2:
[40:25] No.

Speaker 5:
[40:25] It can make its own code.

Speaker 2:
[40:26] Well, that's the whole point.

Speaker 6:
[40:27] It can make a lot of paper clips really fast.

Speaker 2:
[40:30] No. What else might it really be good at finding?

Speaker 4:
[40:33] Oh, vulnerabilities in other people's code.

Speaker 2:
[40:35] Vulnerabilities in existing code.

Speaker 4:
[40:37] So it is its own hacker.

Speaker 5:
[40:39] It'll hack everything.

Speaker 2:
[40:40] Not only its own code, any code, this thing can hack anything.

Speaker 5:
[40:45] Yeah.

Speaker 2:
[40:46] It has found vulnerabilities that have been hiding in core software for years that no other system has been able to detect. And not only can it find it, It's going to wreak havoc. It could reliably exploit them. It's like, oh, there's a vulnerability and I'm going to exploit it. And there you go.

Speaker 5:
[41:06] And damage done.

Speaker 4:
[41:07] And a white hat and a black hat is only distinguished by who is behind it.

Speaker 2:
[41:13] Of course.

Speaker 5:
[41:13] Right. This has no hat.

Speaker 4:
[41:14] Exactly.

Speaker 3:
[41:16] And Thropic, to their credit, they had an oh shit moment where they're like, pull the plug, pull the plug.

Speaker 2:
[41:23] They realized that, first of all, if this gets out, we're screwed.

Speaker 3:
[41:28] Right.

Speaker 6:
[41:29] Oh yeah, they're going to get sued into oblivion.

Speaker 3:
[41:32] Worse than that, Bob.

Speaker 2:
[41:33] I don't mean we, the company. I mean we, civilization.

Speaker 3:
[41:36] Like taking down the entire financial system of the world kind of thing.

Speaker 2:
[41:42] And number two.

Speaker 6:
[41:43] I buried the headline on that one.

Speaker 2:
[41:44] In 12 to 18 months or maybe sooner, everyone's going to be there, right? Because they're only 12 to 18 months ahead of the competition. So what did they do in response to this realization that they now have the ultimate vulnerability finding and exploiting software in the world?

Speaker 3:
[42:02] What did the company do, Steve?

Speaker 5:
[42:03] Put all their money into gold.

Speaker 2:
[42:05] What did the company do?

Speaker 5:
[42:06] Oh, they wrote a new program to stop the old program.

Speaker 6:
[42:10] They invented a time machine.

Speaker 5:
[42:12] They invented an AI to stop the mythic myth.

Speaker 6:
[42:16] We're kind of screwed. Wow. So I mean, what they do is irrelevant, is almost irrelevant.

Speaker 2:
[42:22] They did something that actually is kind of reasonable, right? So they essentially put together a consortium of 40 big giant software companies, including their competitors, pretty much everybody, the big 40 software people in the world. They gave them all the access, the access to mythos review, like a review version of mythos. So it's not fully activated, but it is like a preliminary release of mythos. They call this Project Glasswing, by the way, which is interesting. And it's basically, all right, guys, you've got some time to use this to find and fix the vulnerabilities in your software. You better do it now.

Speaker 5:
[43:04] Oh, so they put on a white hat, basically.

Speaker 2:
[43:06] Basically, they're using it to white hat, basically lock down and secure core software functionality. Now, you got to keep in mind, we think of software, I see we non-experts, experts know this, but we think of big corporations with big software packages, but those are often built upon open source software. The core engines that are driving a lot of even private corporate software is all open source. Open source software is mainly produced and run and maintained by volunteers working on a shoestring budget. If you find a vulnerability in a core piece of software like that, that affects thousands of applications, not just one application. That's at the core of our entire software infrastructure. One of the things they mentioned was they found a vulnerability in the JPEG format. I mean, how common is that? When you could use that as a way to get into any system, you know? Yeah. Holy shit.

Speaker 6:
[44:13] Yeah.

Speaker 2:
[44:14] So anyway, so it's really scary. You have to think about where we are. Now, I'm saying, hopefully, in five years or 10 years, we'll look back at this as like a Y2K situation. You know, those of us who follow this.

Speaker 6:
[44:26] Yeah.

Speaker 2:
[44:26] Like, okay. Remember when that almost happened, but we fixed it by preventing it from happening in the first place. So essentially, we are relying upon what I call the noblesse oblige of tech CEOs. To sort of realize how the incredible power and vulnerability of their own products and to actually do something responsible and proactive about it because no one else is doing it, right?

Speaker 4:
[44:55] Or they just can't keep up.

Speaker 2:
[44:57] The Trump administration has basically decided, and again, this is the same dilemma that you mentioned. Again, you mentioned all the same kind of dilemmas. We're giving it in the hands of many people, which is good, but it's also bad. We've lowered the bar, and it's a powerful tool, which is good, and it's also bad. We want common-sense regulations, but we don't want to stifle innovation, especially with China breathing down our neck. From what I'm reading, China is poised to kick our ass in AI. So I get it. The Trump administration is basically taking a no-regulation approach. We don't want to do anything to stifle AI. In fact, they're trying to prevent the states from regulating AI. They're like anti-regulation, not no-regulation. They're trying to completely deregulate or preemptively not regulate the entire industry, which is not good. We need industry and the government and internationally to get together and say, all right, how do we not destroy civilization with these products? Which gets back to a conversation we've had many times. AI doesn't need to be malevolent or sentient or whatever. It's a disruptive technology. It's disruptive enough and it's moving so quickly that before regulations could be put in place or before we can prevent really bad things from happening, this kind of powerful application can proliferate. We're just lucky in a way that a company like Anthropic, as far as I could tell from everything I'm reading, that are putting on the white hat and going, guys, we should do something responsible about this before this turns into a disaster. You could easily imagine an even slightly less conscientious company being in their position and how much worse this could go. So this is a story we gotta keep an eye on because this is just happening.

Speaker 6:
[46:51] Holy shit.

Speaker 2:
[46:54] This is almost science fiction movie trope level of plot. There have been movies where exactly this happened. Some company develops a piece of software that could basically hack anything in the world and cybersecurity is about to become worthless.

Speaker 6:
[47:11] Yeah, like quantum computers have raised that spectrum for years now. But we've been prepared for that and been thinking about that. And there are things that you can do now with your classical computers to safeguard them against quantum computers wreaking havoc on your encryption. But this is something that could affect, not the super encrypted secret data, but every computer system on the planet could be impacted.

Speaker 2:
[47:36] Absolutely. And again, similar to what Kara was saying, if you look at a cybersecurity, cyber terrorism perspective, we're now giving the tools of cyber terrorism and putting them in the hands of anybody. You don't need a coder. You don't need a world class coder in order to be a cyber terrorist. All you just need is access to an off-the-shelf app. So it's a brave new world we're heading towards with this. We have to keep our eyes open. We have to talk about stories like this. And the powers that be need to get their shit together because they have got to almost Manhattan project this stuff. The consortium is a good start. Hopefully, the consortium will evolve into something more. But we need an international organizations. Like we now, like there's now we talk to, you know, an astronaut about, oh, there's an international organization to look for asteroids which can hit and destroy the Earth. That's a good idea. You know, how about an international consortium whose job it is to look for ways that these emerging technologies could also destabilize our civilization, right?

Speaker 4:
[48:42] Yeah. And I think we also have to be kind of careful from our American perspective often when we talk about these things because we are practiced in existing in a society and a governmental structure that had a fair amount of like democratic morals built in. And so we're often saying like when a nefarious actor, instead of realizing that we may be that nefarious actor.

Speaker 2:
[49:07] Absolutely.

Speaker 4:
[49:08] Like we are definitely at risk of our own government doing some pretty disgusting things with this technology. It's not just them versus us. It's not some boogeyman out there.

Speaker 2:
[49:21] Yeah. I mean, I don't think I said anything to imply that, but it's good to point that out specifically.

Speaker 4:
[49:24] Yeah. We're often like China could and it's like, no, but we could.

Speaker 2:
[49:28] But the point I'm saying, we need to do this, but not only us, this needs to be international. And if we're just looking at who are the top players in AI right now, yes, America is up there, but China is right there too.

Speaker 3:
[49:42] Steve, can their software be used for good, meaning could it figure out all the vulnerabilities that another piece of software has and then give them the ability to fix all of that?

Speaker 2:
[49:53] Yeah, absolutely. That's the point. That's what they're doing. That's their goal of Project Glasswing, identify and fix all these vulnerabilities before this tech gets out there. Because it's happening. Again, 12 to 18 months, probably it's going to be everywhere. If not sooner, you know what I mean? So you better fix those vulnerabilities now. That's the whole idea.

Speaker 6:
[50:17] Yeah, we're going to be hearing a couple of words more in the near future. And those two words are air-gapped.

Speaker 2:
[50:22] Air-gapped.

Speaker 5:
[50:24] Air-gapped.

Speaker 6:
[50:24] Oh, no.

Speaker 5:
[50:25] I don't like the sound of that.

Speaker 2:
[50:26] That's a good thing. Air-gapped basically means that you have a piece of hardware that is not in any way connected to anything, right? So you can't connect. It's not connected to the Internet. You can't access it through Wi-Fi. So it's completely isolated. And therefore, you could have protected information on there that isn't vulnerable for that reason.

Speaker 6:
[50:45] Yeah. And then, of course, that reminds me of the book I read years ago, Robo-Apocalypse, where an AI was so clever, it got around air-gapping in some freaky, cool way. Like, oh, man.

Speaker 2:
[50:57] Yeah. You could use just a flickering light to code or something. Yeah. Interesting. All right. We're not trying to be alarmist here, again, but be alarmed. We're pretty-

Speaker 5:
[51:08] Is it time-dependent?

Speaker 2:
[51:09] We're trying to be realists about AI in every way, right? It's not overhype it, not underhype it, not overworn about it or underworn. This is the company itself. This is anthropic. They're saying, this is a huge vulnerability, and we need to do something about it now or this could be a disaster. They make a pretty compelling case for it.

Speaker 6:
[51:32] I'm thinking, what's the cool side to this? This is powerful hacking, but how is it powerful in other ways that could be dramatic, dramatically awesome?

Speaker 2:
[51:42] When I first read about vibe coding, I'm like, I want to do that. That sounds fun. If I could just make an app for myself that does something, that actually functions, the possibilities there are huge, and I think that is an awesome tool to have. But it just, part of the recommendation is that security and quality control needs to be baked into those apps, not an option, not an add-on, not a premium feature, but absolutely baked in so that you can't, nobody should be able to use one of these sophisticated applications to make code that doesn't already incorporate reasonable security, if not ironclad as much as we could make it. So that's one of the recommendations going forward. But if that's going to be ubiquitous, that has to be regulation. Either that or we're just counting on an entire industry to all do the right thing, and that doesn't seem realistic.

Speaker 6:
[52:41] Steve, what's the government angle here? Because they need to secure their stuff, but maybe even more so than securing my personal computer type stuff. I mean, they have access to computer access to reactors and water filtration and all this huge things. I hope that the government is also intimately involved in this white hatting.

Speaker 2:
[53:05] Yeah. I didn't see that piece of it specifically, but the thing is most of that software also uses the core open source stuff. They will be protected by that as well, but then all the proprietary stuff that they have also needs to be specifically analyzed for vulnerabilities as well. So, gotta hope that happens.

Speaker 6:
[53:29] One of my major concerns is that, yeah, you use mythos to make your software really secure, but then what's that going to do to the user experience of your software and what's it going to break? It's secure as hell, but now it's almost unusable.

Speaker 2:
[53:46] But usually that's not the case, though. These vulnerabilities are like, oh, you could do this clever little hack to get into the back door or whatever, it doesn't necessarily, it's just all back end stuff that I think will be seamless to the end user.

Speaker 3:
[53:58] If you really think about it, their software that can do this is the most important piece of software on the planet, in a way. And it's the most desired, what government or what other software company wouldn't want it, or at least be able to use it on their own stuff. This is profound. This is one of the biggest software-related, technology-related things that I've heard in a long time that...

Speaker 6:
[54:21] Oh, my God. I hope Anthropic, one of the first things they did is to use it to shore up their security. Because you know, when other countries have heard about this, they must have been tried to hack Anthropic just to get their hands on it.

Speaker 3:
[54:34] Right?

Speaker 6:
[54:35] First.

Speaker 5:
[54:35] It's got to be like an arms race.

Speaker 2:
[54:37] We are in an arms race. We are in an AI arms race.

Speaker 6:
[54:40] I hope they used it to secure their own stuff.

Speaker 5:
[54:42] Oh, boy.

Speaker 2:
[54:43] Oof. All right. All right. We'll keep you updated. All right, Bob, tell me about this superconductivity breakthrough. Is breakthrough fair? Is it really? Is it a breakthrough?

Speaker 6:
[54:53] It depends how you look at it, but I'm just having trouble following your talk. Seems kind of less interesting at this point. Okay, let's do this. This was really fascinating. So scientists, is that the first word I use for all of my talk? Scientists or researchers have for the first time, guys directly imaged a new layer of the process that allows superconductivity to happen. So how do they do this? And why do I keep reading about ballroom dancing metaphors to describe it?

Speaker 2:
[55:22] Because of Cooper Paris?

Speaker 6:
[55:24] Maybe. That's a good guess. So researchers include an interesting mix of experimental physicists and theoretical physicists. The former is from the French National Center for Scientific Research in Paris, CNRS. The latter is the Simone Foundation's Flatiron Institute. So, okay, the name of the paper, the paper was published in physical review letters just recently, this April, 2026. The name of the paper got some jargon in here. Observing spatial charge and spin correlations in a strongly interacting Fermi gas. It's much more interesting than it sounds. So, all right, superconductivity. We've mentioned this many times on the show, and I think most everyone is comfortable with what's going on with superconductivity at a high level, electricity flowing without resistance, and it happens in very cold regimes. That's where you're going to see things like superconductivity happening. And do you guys know the key prediction of modern superconductivity theory? What's the key thing? And Steve kind of just ruined that because he basically said it's Cooper pairs, right? You're right, Steve. That's the critical thing right there that's much less much less well known.

Speaker 2:
[56:35] Well, you should have set me up with that dancing part.

Speaker 6:
[56:36] I know. I know, you bastard. So this is called, this is BCS theory named after the physicists who developed it. In the 50s, physicist John Bardeen, Leon Cooper, and John Robert Schreifer. They won the Nobel Prize, as you might imagine. And yes, it's this idea of Cooper pairs of electrons kind of getting together and traveling together. And that being a critical component to what allows superconductivity to happen, electricity flowing without any resistance through wires, whatever it's moving through. So, but this theory, this BCS theory, it's only an approximate framework. I wasn't really aware of how approximate it really is. I mean, it can't describe all the aspects of superconductivity or even all the different types of superconductivity. So researchers have known for years that there is critical stuff missing in this BCS theory. And they've been searching for years and have really hadn't had any major breakthroughs. There's been some developments for sure, but actually understanding it at a much more, at a fundamental level has been kind of very elusive. So this gets us to their new experiment. They used a new imaging method. This involved an ultra cold atom microscope, which is fascinating out of scope for this talk. Maybe I'll talk about it some other day, but they used, using this ultra cold atom microscope, they were examined essentially a Fermi gas. Now in this case, this is a gas of lithium atoms and they cool it to make it into a Fermi gas. They've got to cool it down to a few billions of a degree Celsius above absolute zero. And that always just makes me so impressed. I mean, they're so close. I mean, we all know absolute zero is essentially impossible to reach, but they're within a few billions of a degree. The low temperatures are critical here, right? Because that removes the thermal noise that infests everything that's not this cold, right? So when you get that low in temperature, it allows this group quantum behavior. I mean, these atoms are always quantum beasts no matter what, but once you cool them down and that thermal noise kind of disappears and gets really minimized, it allows this group quantum behavior to take over. And that's what allows things like superconductivity to happen, superfluidity to happen, Bose-Einstein condensates to happen, and of course, Fermi gases to happen as well. You need something to be really, really cold. So this is why, essentially, why electrons will require these really, really cold regimes in order to pair up to form Cooper pairs and do their superconductivity bed. So in this state, then, lithium atoms essentially act like electrons. And it all comes down to the fact that they're Fermions. If you're Fermionic, I don't need to go into detail about that, but they essentially act like electrons. They pair up, they get together, do their thing. So that makes it, when the atoms are pairing up, it turns it into a much more controllable investigation. You know what I mean? So if we can't examine electrons in this way, using this ultracold atom microscope, we need to use these lithium atoms to make it so that it's something that we can investigate in a very controlled, accurate way. So that's why they're using this method. Does that make sense why the guys, you know, they're creating this Fermi gas, ultracold Fermi gas, using lithium atoms, and they act like electrons. They pair, they form Cooper pairs together, and that way we can see what's going on a lot easier than if we were using electrons. So what did they find? After they paired up, after they formed Cooper pairs, these atoms moved in what they described as a synchronized dance. So if you have one pair, if you're looking at one pair, one Cooper pair of atoms or electrons, the position of that pair is actually dependent on the position of other pairs. And that's the key breakthrough right there. So this is where, of course, where the ballroom dancing metaphor crops up. That's why I'm seeing it everywhere I look. Because when you're ballroom dancing, you pair up with somebody and as you're dancing, there's a part of your brain that's making sure that you don't hit other dancers, right? Even though you may be doing complicated moves, you always make sure that you're not going to hit anybody else. So that's kind of what these Cooper pairs are doing. To Tariq Yefsa, the experimental research lead at the French National Center for Scientific Research said, the BCS theory gives us a view from outside the ballroom. So he means here is that the old theory, the way we understand it right now before this experiment, it gives us a view from outside the ballroom where we can hear the music and see the dance come out, but we don't know what's going on inside in the ballroom. Our approach is like taking a wide-angle camera inside the ballroom. Now we can see how the dancers are pairing up and paying attention to one another so they don't bump into each other. So that's his metaphorical description of the advance. So the takeaway here is that this was not predicted by the 70-year-old theory. This is a new insight into what's happening with superconductivity that we did not have before. So in its advances like this, having a better, more fundamental understanding of superconductivity itself at its lowest level, that could lead to whole new industries, essentially. And that's what gets me really excited about this. And that leads me to like, what's the future hold for this? And one of the big takeaways, one of the biggest takeaways from this is that it could lead to the solution to one of the one of the holy grails of modern physics. And what's that? Room temperature superconductors. If we had that, the impacts to our lives and to industries are kind of hard to predict, just how dramatic it would be. I mean, we're talking about ultra-efficient electric grids, the changes to electronic devices, it would be a game changer. I tried to come up with a list of some of the things that you would notice, some of the big, big changes. Like imagine if all the power grids had almost no transmission losses. Imagine if your electric motors and generators were fairly quickly cheaper, denser, and much more powerful because of this. Magnetic levitation and frictionless transport would be ubiquitous eventually, right? Massively stronger magnets, far superior energy stored on the grid. The list goes on and on. It could be extremely dramatic and people have been waiting and hoping for this. But of course, I got to bring in some caveats, right? To be really revolutionary, room temperature itself, just having room temperature S-con cables is not to be all and end all. There needs to be other things need to also happen. It would also need to be cheap. It would also need to be stable. You would also need to have it manufacturable. If it's too hard to manufacture, then all of this means nothing, essentially, except maybe in the lab. It also is going to need to carry very high currents in strong magnetic fields. That might not even be possible. So we don't know. So there's a lot of things that would need to come into place to make this truly revolutionary, as I hope it could be. But if we're ever going to get there, it's a fundamental breakthrough like this that's going to do it. So let's see what happens. That's all I can say.

Speaker 2:
[64:02] Machines operate at higher than room temperature.

Speaker 6:
[64:05] Right. Well, that's true. That's an interesting point. Yeah, because sometimes will they be able to operate at 100 degrees Fahrenheit? That's an interesting point that I haven't really... I can't remember even coming across in terms of superconductivity and room temperature. But yeah, see, that's another thing that's potentially limiting. Even if you have room temperature, doesn't mean, oh, we're going to have all this great stuff. It's a very complicated interplay of all these other characteristics that also need to work.

Speaker 2:
[64:36] Yeah. Usually, technology has to function at a range of temperatures, including much hotter and much colder than room temperature.

Speaker 3:
[64:43] Yeah.

Speaker 6:
[64:43] I mean, even if they got to superconduct at just barely above freezing, that would be amazingly fantastic.

Speaker 2:
[64:50] Yeah. That would be huge.

Speaker 5:
[64:52] Yeah.

Speaker 2:
[64:52] Evan, could superconducting cables be used in red light therapy? Sure.

Speaker 5:
[64:57] Why not? Red light therapy? It's a boundless horizon. Anything's possible.

Speaker 6:
[65:04] The therapy would be cheaper.

Speaker 5:
[65:06] So how about this? From Gentleman's Quarterly, how often do we refer to Gentleman's Quarterly?

Speaker 6:
[65:11] Never on this show.

Speaker 5:
[65:12] Right? I love firsts on our show anytime. How about the title of this article, The Skeptics Guide to Red Light Therapy? How does that not make me slam on the brakes and stop and read that one?

Speaker 6:
[65:24] Better than red light districts.

Speaker 5:
[65:27] Red light therapy, sometimes called photobiomodulation. Yeah, a treatment that uses specific wavelengths of red and near-infrared light directed at the body. Now, it's unlike ultraviolet light, which can damage skin. These longer wavelengths, which penetrate the surface and are absorbed by cells, particularly within the mitochondria. The structure is responsible for producing cellular energy. The idea is that this light exposure, red light therapy exposure, can boost energy production at the cellular level, potentially enhancing repair processes, reducing inflammation and improving overall tissue function. In practice, red light therapy shows up in a wide range of settings and places you will find it. For example, dermatologists will use it in clinical environments to help with skin conditions like acne, wound healing, and signs of aging like wrinkles. But it's also used in treatments for hair loss, where they apply light to the scalp to stimulate hair follicles. Outside of medicine, it's been embraced by the wellness and fitness world, where it's marketed for things like muscle recovery, reduced soreness after exercise, and even improved athletic performance. More recently, it's been promoted for things like chronic pain, and improving your mood, and brain health. So a pretty wide range of things there. Red light therapy can be called a non-drug and non-invasive solution. So you can see why it would have an appeal to many people who for whatever reasons have a distrust of modern medicines and treatments. And red light therapy is very much in fashion right now. It's being touted for all sorts of things. Some of them within the scope of reasonableness, and things really outside on the fringes. But is it some kind of panacea? Hardly. I don't think so. But what do physicians who understand science-based medicine have to say about red light therapy? Steve, you directed me to an article that Scott Gavura wrote for Science-Based Medicine recently about red light therapy. I'll give you a couple of lines from his article there. Proponents claim that this light stimulates mitochondria, the powerhouses of our cells, to produce more energy, ATP, which supposedly improves healing, reduces inflammation and accelerates fat breakdown. How red light stimulates mitochondria is thought to be through interaction with cytochrome C oxidase, CCO for short, which is a key enzyme in the mitochondrial electron transport chain. Bob has nothing to do with the electrons you were talking about. Red and near-infrared light are thought to be absorbed by CCO, displacing nitric oxide, NO, which otherwise blocks mitochondrial respiration. The idea sounds appealing and astonishingly simple. You shine light, energize your cells, and your body performs better. But while this effect on mitochondria has been observed in vitro, very important there, you know, cells in a lab, when you're translating that effect into outcomes that we actually care about, like weight loss or some other measure of improved health, that is a very different challenge. That's what Scott wrote. And that's and this is where the skepticism is definitely warranted. What are the quality of the studies that have been done on people as opposed to, you know, seeing cells react in Petri dishes or test tubes? Well, apparently, it's not all that impressive. You have small studies, short duration trials, manufacturer sponsored trials, relying on less accurate surrogate measures. And the risks of biases, of course, are always high. Do they also look at things like the longer term safety and frequent prolonged exposure to red light therapy? That's not really been well studied. And ultimately, the safety depends on the specific product in use and how it is used. And there are a lot of these products out there. You'll see them in face masks, lamps, wearables, all sorts of... I've seen... they have pictures of dogs. You sit your dog in front of a red light for red light therapy. And there are thousands, actually, of these products that get marketed. So, you know, does it have a real scientific basis to it? Yeah, you know, maybe it does, but how does that translate to all the uses that red light therapy claims it can help with? It depends. Are you using it to help with your skin condition? Okay, dermatologists will do that. But are you using the same red light therapy products to help lose weight, stimulate muscle growth? It becomes a problem with devices or techniques that have limited actual scientific underpinnings to it. Right, you have this kernel of truth upon which is built this whole market place basically for people seeking these non-invasive and non-drug treatments. And the GQ article actually I thought did a pretty sensible take on it. It was not a bad approach. It treated red light therapy as something that has limited benefits, skin, hair, and maybe a little inflammation, maybe some muscle recovery. But again, it acknowledged that a lot of the evidence is early, it's fragmented and definitely, it's definitely oversold. Red light therapy is not magic. It's not one of those rare treatments that somehow improves everything in your life, you know, including workout, brain fog, and all sorts of claims. You're not going to get those kinds of results from these things that you're picking up at your local mall. And there's plenty of studies that have looked into this. I'll give you a couple of examples. Stanford Medicine says the evidence is fairly robust for hair regeneration and wrinkle reduction. But the other applications, such as enhanced athletic performance, fat reduction, fat burning, they describe it as less well-supported. That means it's wanting. Even the American Academy of Dermatology, where dermatologists say, yeah, we use red light therapy, but it's a complementary therapy. It's not a replacement for standard treatment. It's just something that you use in addition to other treatments. How do you really suss that out of what is doing the work here? Where are you getting the real results from? Is it from the red light therapy product, or is it from the other things that your doctor is asking you to do for yourself? These home devices are so wild and variable. Different products, how much light are they putting out? What are the doses? What are the protocols? There may be a signal there, but there's a standardization problem. One mask that you're wearing is not necessarily like another mask. And of course, whatever you're being told, they're going to oversell it. They're going to hype it. And Scientific American recently noticed about that very problem. It's the hype machine, right? They say they've got something a little plausible here, but the enthusiasm is racing ahead of the clinical evidence. It was a 2026 study on how at home red light therapy devices are promoted on social media, of course, and explicitly framed those promotions in the context of that limited evidence. And we see this a lot. It's all over wellness culture. You take a real mechanism, you add some small studies, strip away all the caveats, and then you market it as some kind of life hack or life upgrade. But before long, the modest possibility that something may help a little becomes almost a certainty that it will help a lot. There's the hype machine for you, overselling what it is actually doing. So we have every reason, I think, to be skeptical, lots of reasons to be skeptical about red light therapy, except for some maybe very specific uses.

Speaker 2:
[73:19] I don't buy the whole mitochondria thing.

Speaker 5:
[73:21] Even that?

Speaker 2:
[73:22] I think it's a weak mechanism. It's not very plausible. The reason is that energy production is incredibly regulated within cells because it's a massive trade-off. It's not like more energy is always better. There's no downside to it. When mitochondria make more ATP, that is like the main metabolic stress on cells. You don't want to make more energy than you need. You're taking a homeostatic, carefully regulated, very complex system, and you think you're just going to boost it in one direction, that's going to be an unalloyed good. It's nonsense. That never works out. Any of these claims that this is going to boost this aspect of biological function and it's going to be a good thing, that almost never works out.

Speaker 4:
[74:07] Or it has a downside to it.

Speaker 2:
[74:09] Yeah, but there's trade-offs. You're going to age faster though too, because that's what happens when it turns into...

Speaker 5:
[74:15] Oh, I see.

Speaker 4:
[74:16] I also don't get how it's... It's only talking about skin cells though, because how does it get to any other cells?

Speaker 2:
[74:23] Well, some frequencies of light can penetrate deeper into tissue than others, and the longer wavelength, redder frequencies can penetrate more.

Speaker 4:
[74:31] But how deep? They're not getting down to your organs.

Speaker 2:
[74:34] Well, you would need near-infrared light for that.

Speaker 4:
[74:36] Right, which would be dangerous for other reasons.

Speaker 2:
[74:39] I guess to the heating tissues.

Speaker 5:
[74:41] Right, and I don't think the products you're buying in the mall are doing that.

Speaker 4:
[74:45] No, I think it's just like a red light bulb.

Speaker 5:
[74:47] Basically.

Speaker 4:
[74:48] Yeah, so I think it's just reaching like the top layer. Maybe it's penetrating a little bit into more dermal layers, but beyond that, I don't see how it could have any claims for anything beyond healing superficial wounds, and even then, I don't think it's feasible. But I just don't even understand the mechanism of how it would get to an organ system, for example.

Speaker 5:
[75:08] And it's interesting what you brought up, Steve. Is there a condition by which a person could have an ailment in which their cells are not producing enough?

Speaker 2:
[75:17] Well, again, if you're talking about treating something that's out of whack, that's different than boosting a healthy system to be even more effective. Those are two very different claims, right? So if you were saying, oh, there's a condition where people, their mitochondria are dysfunctional and they can't produce enough ATP, and we're going to compensate for that by boosting their production with red light, I would buy that chain of argument.

Speaker 5:
[75:43] But they're not saying that.

Speaker 4:
[75:44] But that's what the wellness industry does. They make everybody think that their cells aren't working well enough. They make everybody think that they have some sort of failure inside of their body that they need to optimize. But you're right, Steve. It's the same as all the claims we hear, all the time, about boosting your immune system. You don't want that unless you are immune deficient. You do not want a boosted immune system. Exactly, right, because then you have an autoimmune problem.

Speaker 2:
[76:11] Even the antioxidant thing is wrong, too. It's the same thing. It's a homeostatic system. Let it do what it does. Unless you have an abnormality, a disease of it, you don't want to mess with it.

Speaker 4:
[76:21] And it's the same thing with vitamins and supplements.

Speaker 5:
[76:24] That's what I was thinking as well, right? Unless you have scurvy or something and need more vitamin C.

Speaker 4:
[76:29] Yeah, some of us...

Speaker 5:
[76:30] Don't hyperdose on vitamin C.

Speaker 4:
[76:32] And do it based on your numbers. Don't just assume that you can take 10,000 times the dosage of something and that more is better. It does not work that way.

Speaker 2:
[76:42] I always check my patient's vitamin levels before supplementing.

Speaker 5:
[76:46] They seem to have studies, though, that say it's good for wrinkles. I mean, that's what they pointed to.

Speaker 4:
[76:51] It's hard, too, because a lot of skincare studies are based on self-report. And it's a really hard thing to say, like, is your hair fuller? Does your skin look, you know, less wrinkly?

Speaker 5:
[77:03] And what else are they doing at the same time? What other actual topical creams and stuff are they putting on to their body?

Speaker 4:
[77:11] I know, like, putting moisturizer on your face makes your wrinkles less apparent.

Speaker 2:
[77:15] Noticeable?

Speaker 4:
[77:16] Yeah, exactly.

Speaker 2:
[77:17] All right, Jay, it's Who's That Noisy Time?

Speaker 3:
[77:19] All right, guys, last week, I played This Noisy. Now, I was impressed because a lot of people got this right. Of course, I usually pick the first person that sends it in. So let's go through this. Listener name, Mike Johnson said, Jay and everyone, my guest for this week's Noisy is a hydropower grain mill. Thank you for all that you do on this podcast and elsewhere. And they always look forward to the banter in analytical and critical thinking each week. So a hydropowered grain mill is one of those mills that is powered by water flow, right? They have that large wheel on the outside of the building. And that wheel turns like a stone grinding disc inside of the building. And that's where they would do. That's how it's powered. You guys can visualize that, right? That is not correct. But that is the first time anyone has ever guessed that specific thing. I think that's really fun and cool. Another listener named Dan Lee wrote in and said, Dear Jay, this week's Noisy is some people doing sprints in a gymnasium or playing racquetball. I can't spell racquetball and I am too sleepy to look it up. Maybe I should have said pickleball. I can spell that. Well, wouldn't your email program spell correct it for you? Or are you right? I'm just curious. But Dan, thanks for sending that in. You're not correct, but after re-listening to it, I don't think it's a bad guess at all. I had a couple of people have close guesses. The first one was David Sear. It sounds to me like someone's standing under a raised subway track and the weight of it going by is making a support or something squeak. That is a very good guess, although it's not it's not correct. And then Shane Hillier. Shane Hillier wrote, it sounds like the BART Bay Area rapid transit when it goes under the San Francisco Bay. Right. So you guys are getting a better idea what this thing is. I did get someone who nailed it. This is Corey Belak and Corey says, long time, first time. I think this week's noisy is of cars driving over an expansion joint on a bridge. Specifically, the expansion joint is made up of two or more metal plates, which overlap each other as the cars drive by or over the joint. The joint flexes a little bit, which causes the plates to rub against each other and makes a squeaking noise.

Speaker 5:
[79:44] I thought you were going to say, specifically, they identified which bridge, which highway, which exactly?

Speaker 3:
[79:50] Well, the previous guesser said it was that BART Transit.

Speaker 5:
[79:56] Like San Francisco?

Speaker 3:
[79:57] No, that's not the correct thing, but someone did pick a very specific thing, and I guess where they live. So anyway, that is correct. So let me play this for you again. Yeah, and that's basically the expansion joint squeaking as the cars go over.

Speaker 2:
[80:20] Yeah, once you know it, like yeah, of course, I've heard that a thousand times.

Speaker 5:
[80:23] And it's designed, right, to do that. In other words, expansion make, there is supposed to be that sort of gap in a sense.

Speaker 3:
[80:30] Absolutely, yeah, that gap allows, like in the winter or the summer, when the size of the bridge changes due to temperature, that expansion joint allows it to move.

Speaker 5:
[80:39] And it's perfectly safe.

Speaker 3:
[80:40] Oh, hell yeah. Yeah, I mean, if you didn't have them, the bridge would eventually break from the temperature changes.

Speaker 5:
[80:47] That's what I want to hear.

Speaker 3:
[80:48] Yep, it's good. That's a good thing. All right, I have a new Noisy for this week. This was sent in by a listener named Matt Tinnenberg. And check this out.

Speaker 2:
[80:59] For life requires that, requires a tradition that affirms it. The Catholic, as I try to spell out in the book, and it's sort of not even hot.

Speaker 5:
[81:17] At the end of every year, I write a piece on for...

Speaker 3:
[81:22] This might be one of the weirdest noisies. I mean, once you find out what it is, it'll make perfect sense. I think it's gonna be very hard to guess what it is, but I'm gonna just straight up tell you, there are different voices in there, and there's properties about those different voices that you should pay attention to. And I think you're really gonna like it. It has a very funny element to it, to me personally. I think it's kind of wacky and funny.

Speaker 5:
[81:47] It was a little disconcerting, but okay, good.

Speaker 3:
[81:49] So, if you think you know what this week's noisy is or you heard something cool, email me at WTN at theskepticsguide.org. You guys know there's a ton of events that the SGU are doing. You can go to theeskepticsguide.org to check out the varied list of events. I'm not going to list them again on the show, but I just want to put a highlight on our Nauticon Conference, which is happening in Australia. This is going to be happening in July of this summer, depending on where you are in the world. It could be winter. The exact dates are going to be July 23rd to 25th. The exact dates are July 23rd to 25th. There's lots of different tickets to be bought at that specific conference. If you're interested in learning more, you can go to nauticoncon.com or you can go to skepticon.org.au. The other thing that we'll be doing when we're on that trip is we will be doing a conference in New Zealand and you can get more information about that conference at conference.skeptics.nz.

Speaker 2:
[82:55] All right. Thanks, Jay. We're going to do a couple of quick emails. First one is just a correction. A couple of weeks ago, we were talking about how fast the Artemis 2 was going. I believe at some point I said it was going Mach 32. You did?

Speaker 5:
[83:12] You were mocked for that.

Speaker 2:
[83:13] I was mocked. I did not know this, but you guys know what the technical definition of Mach speed is?

Speaker 3:
[83:19] Yeah.

Speaker 6:
[83:20] The speed of sound at sea level?

Speaker 3:
[83:23] No, but if you go like Mach 2, you're going twice the speed of sound. Mach 3 is three times the speed of sound.

Speaker 2:
[83:29] Yeah, but you're missing a detail. Which speed of sound?

Speaker 3:
[83:32] It's going to be at sea level, of course.

Speaker 2:
[83:34] No. No, that's the mistake we all make in a medium.

Speaker 3:
[83:37] Well, I'm wrong then.

Speaker 5:
[83:38] To do with the medium?

Speaker 4:
[83:39] In air?

Speaker 2:
[83:41] Mach is a unitless ratio, defining the object speed relative to the speed of sound in the surrounding medium.

Speaker 5:
[83:49] There you go.

Speaker 3:
[83:50] Yeah, it's relative shit.

Speaker 2:
[83:51] When you're in space, there's no Mach, right? Saying the Artemis was going Mach 32 while outside the atmosphere is wrong. Although, that's what I was just quoting the sources that I was reading, so I guess it's a common misunderstanding. You could say it was going 32 times the speed of sound at sea level. That's fine, but you can't say Mach 32. Does that make sense?

Speaker 6:
[84:14] Nice.

Speaker 2:
[84:14] Nice wrinkle.

Speaker 6:
[84:16] Interesting.

Speaker 2:
[84:16] Unitless ratio. That's cool. I did not know that. Thanks for sending that in, those who did. All right, this one's interesting. This comes from Adam Pesci. He said pronounced like Joe. I'm assuming he means Joe Pesci.

Speaker 5:
[84:27] Well, maybe they're related.

Speaker 2:
[84:29] And Adam writes, the talk around my Tuesday morning pickleball games often revolves around conspiracy theories like aliens on Earth, UFO talk, et cetera. I haven't seen concrete evidence to take any of this seriously, although any help from aliens would be welcome. This week, the discussion was around the missing or strange deaths of US scientists tied to NASA, JPL, or nuclear space departments that a ninth scientist was just announced to passed a few years ago. I've seen reporting from the Daily Mail and most recently this week on Fox News. Not exactly reliable outlets. My skeptical antennae are raised. However, the numbers do seem higher than coincidence. I would love to get your take. This subject touches on both science and critical thinking regarding conspiracy theories. Love your work. Thanks for all that you do. Have you guys seen this?

Speaker 3:
[85:19] No.

Speaker 5:
[85:19] Yes. I've heard of it. No.

Speaker 2:
[85:21] It's going around social media. I think there's one or two more have been added in the last week since we got this email. There's at least nine now on the list. I looked into it as much as I could. There isn't a lot of specific information on it. It seems like the individual cases are all true, although I've looked at these kinds of conspiracies before, meaning the mysterious deaths of people, and they're not always that mysterious. You know what I mean? The mystery is being ginned up. The same is largely true here, although missing is missing. There are some people who just vanished, who just went missing, and at least one of the people was murdered, like it was a murder. Carl Grillmayer, a Caltech scientist, shot dead on his porch on February. Sometimes the elements are misleading, and in this case, it looks like some of the deaths were just deaths, some were violent, a couple of people were shot dead, and some people are missing. But the question is, could this all be a coincidence? And so I think that is the default answer that would need to be rejected before you would leap to any notion that there's a conspiracy going on. Of course. Yeah, of course. And what people say, nine scientists. So first of all, they do another thing that they inflate the numbers by, first of all, going back in time. Well, how far back are we going to go? They just added somebody from 2023.

Speaker 5:
[87:02] Yeah, keep going until you get some kind of result that you like.

Speaker 2:
[87:05] Yeah, so that's at least over three years now, right? It's between 2023 and 2026.

Speaker 4:
[87:10] How many scientists are there?

Speaker 2:
[87:11] That's the other thing. What's the baseline? How many scientists are there? Is this unusual for the number of people that you could potentially be drawing from? And so that's the other thing. It's not like all these people worked in the same lab, right?

Speaker 4:
[87:26] Right.

Speaker 2:
[87:26] They're pulling in anyone who might possibly have any kind of relations. Like one person is a retired Air Force Major General. Okay. How does that relate to a lab worker from Los Alamos National Laboratory? He's not even a scientist, right? He's a lab worker. Another one is a laboratory employee, right? Again, not even a scientist.

Speaker 4:
[87:49] How many desks were there at Google over the last five years?

Speaker 2:
[87:53] They're throwing the widest net. One is MIT science, plasma scientist. So how does that relate to Los Alamos? How does that relate to a retired Air Force Major General? So they're spreading it out a lot. It's like saying celebrity deaths occur in threes, but you're really stretching the definition of celebrity.

Speaker 5:
[88:13] Cast the widest net possible and you'll catch everything.

Speaker 2:
[88:16] Exactly. So they're catching a very wide, it's almost like a mentalism game. They're casting a deceptively wide net, over a period of time, and stretching the definition of mysterious as well. So that's called subjective validation, when you're able to play with the definitions enough to make a pattern appear when there isn't really one. The only thing I would say, so I think that's enough to explain this, but the only thing I would say is that, is it possible that one or more of these people were bumped off by international spies who were trying to, whatever kill or maybe they were spies and they got bumped off by their handler because they were selling secrets in North Korea? Who knows? I don't know that there isn't anything actually nefarious mixed in here because again, it hasn't really been long enough for people to do investigative journalism type stuff on each individual case and the government's not saying anything.

Speaker 4:
[89:13] Right, but base rates, what is more likely that somebody died by suicide?

Speaker 5:
[89:18] Statistically, what it was.

Speaker 4:
[89:19] Exactly.

Speaker 2:
[89:19] I'm just saying, we can't rule out that there isn't anything nefarious going on buried in here. But the overall claim of these nine scientists are connected, they're not that connected and that they're missing or their deaths are too unusual to explain by coincidence, just isn't true. It absolutely could simply be a coincidence. It could be the lottery fallacy. It's like somebody noticed this and then they went looking for other instances to bulk up the numbers. This is exactly what you would expect by chance in similar situations. So not impressive.

Speaker 5:
[89:54] You like an allegory?

Speaker 2:
[89:55] Sure.

Speaker 5:
[89:56] Gerald Posner, the author of the book Case Closed, who did a very deep dive into the JFK assassination. He has an appendix in his book titled The Non-Mysterious Mystery Deaths. And here's what he recently wrote, a quick paragraph. For Case Closed, I spent months investigating 101 mystery deaths, air quotes, JFK theorists claim provided key witnesses that had been silenced. Not one of them held up under scrutiny, not one. Sensational headlines, zero real examination. So he said that and that is an investigative journalist who actually did that work on 101 of them in this particular case, and they were all explainable.

Speaker 2:
[90:38] Yeah, so it would be interesting to see what that kind of investigative journalism reveals about this case. I'm speaking from skeptical first principles, like wasn't able to find much in internet research in terms of the specific cases. But the older conspiracies, yes, usually like if someone's talking about something over 20, 30 years ago, it's like usually that they've been completely sussed out.

Speaker 5:
[91:00] It susses out.

Speaker 2:
[91:01] Yeah. All right, guys, let's go on to science or fiction.

Speaker 1:
[91:07] It's time for science or fiction.

Speaker 2:
[91:17] Each week, I come up with three science news items or facts, two real, one fake, and I challenge my panel of expert skeptics to tell me which one is the fake. Are you guys ready for this week? All right, I think we got three interesting news items this week. Here we go. Item number one, a comprehensive review finds that stingers, horns, and teeth across a broad swath of the plant and animal kingdoms share a similar shape largely due to remarkable convergent evolution. Item number two, a new technique can simultaneously improve the strength and ductility of titanium alloys in just milliseconds using less than 50 percent of the energy of heat treating. And item number three, researchers have put the final nail in the coffin of MOND, that's the modified Newtonian dynamics, by demonstrating stable gravitational force laws at vast intergalactic distances. Jay, go first.

Speaker 3:
[92:18] The first one, a comprehensive review finds that stingers, horns, and teeth across a broad swath of the planet and animal kingdom share a similar shape largely due to remarkable convergent evolution.

Speaker 2:
[92:30] Basically anything with a point, you know?

Speaker 3:
[92:32] They've come to be in similar shapes because of convergent evolution, and not just because it works.

Speaker 2:
[92:40] Well, that is convergent evolution.

Speaker 3:
[92:41] That's what I'm saying.

Speaker 2:
[92:46] It's functional, because it's functional, the selective pressures lead to the same outcome.

Speaker 3:
[92:52] I think this is totally science. I mean, imagine asking people to, you have boats and you're like, build something to push into the water so you can move forward. I think there'd be a lot of things that look very similar to modern paddle, right? I mean, I think it makes perfect sense. Second one, a new technique can simultaneously improve the strength and ductility. What does that mean?

Speaker 2:
[93:14] Like the ability to stretch like taffy without breaking.

Speaker 3:
[93:18] All right. So to improve the strength and ductility of titanium alloys is just milliseconds using less than 50% of the energy of heat treating. It certainly seems like a big claim, but okay. All right. I don't know about that one. Last one, researchers have put the final nail in the coffin of Mond by demonstrating stable gravitational force laws at vast intergalactic distance.

Speaker 2:
[93:42] We've talked about this a lot in the show. You're going first, so I'll give a little bit of background to this because that word Mond is doing a lot of heavy lifting in that sentence. Remember, that's modified Newtonian dynamics. This is the alternate hypothesis to, Bob?

Speaker 6:
[93:57] Dark matter, man. It's obvious.

Speaker 5:
[93:59] I mean, dark matter, exactly.

Speaker 6:
[94:00] Jesus. It was such a silly question. I thought you were joking.

Speaker 5:
[94:07] What's two plus two, right?

Speaker 2:
[94:08] I wanted you to say it. I'm just trying to be interactive here, Bob, right? And so the idea is that Newtonian dynamic, the Mond hypothesis that, no, we don't need dark matter, but gravitation behaves differently at different scales. That's Mond, basically. This is saying, nope, Mond, is proven false because gravitational force lowers are the same even across vast intergalactic distances.

Speaker 3:
[94:34] All right, so I'm going to first say, who the hell am I to even comment on this? Right, because, you know, I'm not an enthusiast like Bob is, so this is like completely based off of nothing other than what's in my head right now without any like empirical understanding or nothing, right? I'm just guessing here. So this is what my skeptic, my critical thinking guess is going to be on this, that gravity functions the exact same way, no matter how small or far the distance is, everywhere in the universe at the same time. So that one, apparently, I'm agreeing and saying that this is, that one is science as well. So the obvious one here that seems crazy is the one that's making this millisecond claim about strengthening titanium, both in its overall strength, because there's different types of strength, as we know, which you didn't specify, and its dulcetility.

Speaker 4:
[95:33] Ductility.

Speaker 3:
[95:34] Ductility, right there. It's very ductile. Very dulcet. The thing that bothers me in this position, particularly because I'm going first, is that I could clearly see that Steve would be playing with this one, meaning it sounds like he's claiming too much. It's the perfect one to put in there as the trick, and it's true. But the problem is that the other two items to me seem perfectly reasonable. So I'm just going to go with my gut and say the one about ductility is fake. It's not true. Don't believe it.

Speaker 2:
[96:05] Okay, Evan.

Speaker 5:
[96:06] All right. The first one about the comprehensive review finds that stingers, horns, teeth and toes, teeth and toes, across a broad swath of the plant and animal kingdoms share a similar shape. Okay, I believe this to be true as well. Boy, I mean, why not? It's such an effective shape.

Speaker 2:
[96:32] Shape? Thank you. Orphology.

Speaker 5:
[96:35] Thank you. You know, that, why, sure. Yeah, convergent evolution, absolutely. Of the three, I'm probably most confident about that one. The second one, about the technique that simultaneously improves strength and ductility. Perhaps the simultaneously part here is would be the fiction on this one. Couldn't tell you which of the two, maybe. It could be one or the other, or maybe it doesn't happen simultaneously. In just milliseconds, snap your fingers, Thanos kind of thing, and it's done. I'm skeptical of that. And Mond, I didn't hear Bob talk about it yet, so I would reserve myself there. I'm leaning towards what Jay is thinking, and I will join him in his skepticism of the Titanium Alloys being fiction.

Speaker 2:
[97:28] Okay, Kara.

Speaker 4:
[97:29] Of course, you're going to call on me next. I don't know anything about the Titanium Alloys or Mond. So that is, it's so hard when you kind of have some understanding of some of the stuff for one of them, but not for the other two, because that can trick you in both ways. Because I'm not confident that stingers, horns, and teeth are due to convergent evolution, mostly because I don't think of those as like fundamental evolutionary properties. Are they that old? Stingers, horns, and teeth, they feel like adornment or like new techniques for capturing prey or for eating or for fighting, or sometimes just display, only display. And that kind of stuff I think usually kind of evolves out of control after time. But maybe it's one of those things like the circle, you know? It just works in nature. So we see it over and over. A new technique simultaneously improving the strength and ductility of titanium in just milliseconds using less than 50 percent of the energy of heat treating. Okay. Is that a crazy claim? You're going to be like, this is amazing when you tell us that it's science. Okay. And then the fact that the MOND thing, you're like the final nail in the coffin. I mean, does that mean that nobody believes it? There's always going to be some diehards, right, who dedicated their whole career to this.

Speaker 2:
[99:04] That does not imply that there isn't somebody's outlier somewhere who's going to cling to MOND. Exactly. Scientifically speaking, like, like it's not holding water. This was the one piece to the puzzle that we were waiting for to say, that closes the door on MOND.

Speaker 4:
[99:17] Okay, so, stable gravitational force laws at vast intergalactic distances. I'm just going to say, I think we live in a relativistic universe and we're not going to see something very, like, Newtonian like that. Like, we're not going to see the same, you know, these stable laws at all these different places where there's, like, massive differences in mass, basically. So, I don't know, I'll say that one's science and I'll say that the Titanium Alloy science, because I don't know enough about it. I'll go with the Stingers. Is that just me?

Speaker 2:
[99:49] Yes.

Speaker 4:
[99:50] Oh, so you're not swept?

Speaker 2:
[99:51] Well.

Speaker 4:
[99:51] Oh, you could be. Okay.

Speaker 2:
[99:53] Could be.

Speaker 4:
[99:53] Gotcha, gotcha.

Speaker 2:
[99:56] All right, and Bob.

Speaker 6:
[99:58] Yeah, no, I'm not picking the Mond one. That makes absolute, the Mond one, number three here makes a lot of sense. Dark, you know, Dark Matter always had scenarios where Mond really had no say, right, Steve? The Bullet Cluster, there were things that were evidence for Dark Matter that just Mond, I think, couldn't touch. It's like, well, I'll explain that then if Mond is a thing. It just didn't make any sense. The only arena where Mond may have had some, something important to say was these vast distances and gravity might be, you know, the laws of gravity might change in not insignificant ways when you're dealing with thousands of light years or at the galactic scale or beyond. So I'm not, I don't have a problem with that at all. The other two that are tough, the Stingers and Horns and Stingers, yeah, it could be convergent evolution potentially. I mean, they basically need to pierce something that could be tough. You know, they're piercing something generally, but then Kara makes a good point about, you know, like horns could be, could have, you know, eventually changed to a sexual display. But I wouldn't be surprised if they had, you know, also important piercing function. I don't know. So this one's tough. And then the alloy, this titanium thing is interesting, but I mean, heat treating metals, complex process, there's a lot of components to it. I'm not sure less than 50% of the energy of heat treating them. So I assume that means, you know, uniformly heating up the metal doesn't have to be done. That energy wouldn't need to be done. So what does that mean then? Sure, you could be doing something sneaky, because number one here with the horns and stingers and teeth sounds much more reasonable than this heat treatment one. But I'm gonna go with Kara and just say there's something about that is not convergent. And I could be wrong because my knee-jerk reaction here is that is this heat treatment is out of whack. But I don't know enough about it to comment really thoroughly.

Speaker 5:
[102:04] All right, it happened last week, didn't it? Me and Jim.

Speaker 2:
[102:07] You guys are divided between the first one and the second one. But let me say something about each of them, though, before I do the reveal. Because I don't know that any of you hit upon the key piece of any of these. So the thing with MOND, it's not so much about whether or not you believe MOND can't be true or not. It's whether or not you think we have a way of measuring gravitational force laws at vast intergalactic distances, right? That's really the key claim in that item. Well, that's the question. And if we do, why haven't we done it until now?

Speaker 6:
[102:39] Well, that is a valid question. What's taken so long to do it? This theory, this experiment apparently has.

Speaker 2:
[102:46] The real key piece of the heat treating one, the Titanium Alloy one, is simultaneously improving strength and ductility. That's what I said. As Evan said, because those are usually properties that are a trade-off. Yeah, that's true. As you improve one, you reduce the other one. Well, crash.

Speaker 6:
[103:01] That's true.

Speaker 2:
[103:02] We have talked about this before, but in situations that can improve both are always great because you're not, yeah, you're not having to do the trade-off.

Speaker 6:
[103:11] It's not impossible by definition.

Speaker 2:
[103:13] The first one about the Stinger's horns and teeth is if it's not at converged evolution, then why do they look so similar? Why do they have that same conical shape at the end? Well, that's what you need to think about.

Speaker 6:
[103:26] Intelligent design?

Speaker 2:
[103:27] All right.

Speaker 5:
[103:28] I wasn't going to say it, Bob.

Speaker 2:
[103:31] Let's go, we'll start with number three. Since you all agree on that one, researchers have put the final nail in the coffin of Monde by demonstrating stable gravitational force laws at vast intergalactic distances. You all think this one is science and this one is science. It is science. So two of you are okay. So, Bob, what did I miss? Massive bonus points. If you could tell me what technique they used to measure the gravitational force law in cosmological scales. I'll give you partial credit if you just tell me what data it involved. Because when you're measuring anything at cosmological scales, what information are you usually using?

Speaker 6:
[104:12] Well, it's got to be-

Speaker 2:
[104:16] Say it.

Speaker 5:
[104:16] Background radiation.

Speaker 2:
[104:18] Yes, the Cosmic Microwave Background.

Speaker 6:
[104:20] Right. Avin, way to go, man.

Speaker 2:
[104:22] Good, Avin. So they combined data from the Cosmic Microwave Background intensity maps from the Atacama Cosmology Telescope and a Galaxy catalog to measure, there's no way, this is like two levels of expertise beyond us, the Kinematic Sunyav-Zeldovich effect.

Speaker 4:
[104:41] Of course, that, the Zeldovich, yeah.

Speaker 2:
[104:44] And I had to look up what that was. It's a subtle distortion in the Cosmic Microwave Background caused by the motion of Galaxy clusters or other large scale structures. So they use that effect to measure the motion of these large Galaxy clusters, and then use that to measure the force of gravity that was in effect at that scale. And what they found was that it's exactly what we predicted to be. It's exactly the same as it is at both Newtonian and Einsteinian, right? Both relativistic and classical Newtonian gravitational effects match what we measure at any other scale. So again, this is one paper, you always have to, if this holds up, et cetera. But this data being true pretty much is the final nail in the coffin of MOND. It closes the door, as they say, because that was their holdout. We don't know that gravity isn't different at cosmological scales. Well, yes, we do. It's the same.

Speaker 6:
[105:43] Steve, I would not have guessed that the cosmic microwave vacuum radiation would be used in that determination. That's not obvious to me. Oh, yeah.

Speaker 2:
[105:53] Yeah, it's not. I agree. I agree. That's why I said you get bonus points for that.

Speaker 6:
[105:57] That's a surprise. Good work, man, because that would not have been on my top five guesses at all.

Speaker 5:
[106:03] We can stop right here.

Speaker 2:
[106:05] Let's go back to number one. A comprehensive review finds that stingers, horns, and teeth across a broad swath of plant and animal kingdoms share a similar shape largely due to remarkable convergent evolution. Bob and Kara, you think this one is the fiction. Jay and Evan, you think this one is science. This one is the fiction.

Speaker 4:
[106:27] Yeah, baby.

Speaker 2:
[106:28] This is the fiction. Good job, Bob and Kara. If it's not convergent evolution, what is it? Because the study showed that something else is largely responsible for the shape, the similar shape of Stinger's horns and teeth.

Speaker 4:
[106:45] When you say something else, is it an evolutionary principle?

Speaker 2:
[106:47] No.

Speaker 4:
[106:48] Okay.

Speaker 2:
[106:49] Nothing to do with evolution. Environment.

Speaker 4:
[106:51] Yeah. Well, that is kind of evolution. Well, it would be then if it wasn't like having that trait and passing it on, then it would be that the environment is shaping those things like more Lamarckian.

Speaker 2:
[107:04] So you're still thinking kind of, well, shaping them how?

Speaker 4:
[107:07] Like physically, like morphologically. Yeah, you're right.

Speaker 2:
[107:10] So here's the title. The geometry of nature's stingers is universal due to stochastic mechanical wear.

Speaker 4:
[107:20] Oh, so it is kind of like circles.

Speaker 2:
[107:22] It is. It's the mechanical wear on them. And how did they show this? Because they showed that over age, these features take on that shape through wear and use. They don't have that shape at the beginning.

Speaker 4:
[107:36] Oh, nature's pencil sharpener.

Speaker 2:
[107:38] They use pencils as an example.

Speaker 4:
[107:39] Oh, that's funny.

Speaker 2:
[107:41] It's nature's pencil sharpener. Absolutely. Pencils have the same shape. Pencils have the same shape. So yes, they mathematically defined that roughly conical shape. The slope increases with the radius, it's proportional to the radius, etc. That gives you sort of that pointy shape that you see with like horns and stingers and thorns and teeth or whatever. And they showed in multiple examples that those shapes develop with use and wear. They're not there without use and wear. And so it's a weathering process, what they call stochastic weathering process. It is not entirely, that doesn't mean there's no role. That's why I had to say largely, there's no role for evolution, but it seems to be largely due to this stochastic wear, not genetics.

Speaker 6:
[108:32] So Steve, juvenile bees with stingers have stingers that are different than adult stingers?

Speaker 2:
[108:37] That's what they're saying in this study.

Speaker 4:
[108:38] Yeah, maybe they're just wider.

Speaker 6:
[108:40] Well, what about bumblebees that disembowel themselves when they sting? I mean, so what's their stinger like? Cause they only use it once.

Speaker 4:
[108:47] Yeah, but that's only one animal. They're talking about the entire animal and plant kingdom. There's probably some conversion.

Speaker 2:
[108:55] As I said, it doesn't rule out evolution at all. But they're saying that it seems to be dominantly wear and tear, not convergent evolution.

Speaker 4:
[109:03] Yeah, that's interesting.

Speaker 2:
[109:04] Yeah, interesting, right? Okay, that means that a new technique can simultaneously improve the strength and ductility of titanium alloys in just milliseconds using less than 50% of the energy of heat treatment is very cool material science.

Speaker 6:
[109:19] What's the deal?

Speaker 2:
[109:20] What's the what is we are? Give me throw something at me. What do you like? It's not heat. It's not light vibration, not light, not vibration, sandblasting chemical treatment. No, not chemical, not sandblasting.

Speaker 4:
[109:33] Electricity.

Speaker 2:
[109:34] Electricity. Oh, it is a massive electric current. That's what they use. That's why it only takes a millisecond. It's like, boom, just shock the hell out of and then what?

Speaker 6:
[109:46] And that takes the place of heating?

Speaker 2:
[109:47] Yeah. So what does heat treatment do? Like just in a broad, you don't have to give me too specific. Just in a broad sense, what does it do to the metal?

Speaker 6:
[109:55] It hardens it. Yeah.

Speaker 2:
[109:57] How? How?

Speaker 6:
[109:59] It realigns the grains of the metal, right?

Speaker 2:
[110:01] Yes. It affects the micro structure of the molecules of the alloy, right? The grains of the crystals of the alloy. So it's affecting the architecture.

Speaker 5:
[110:13] Like a snap to attention.

Speaker 2:
[110:15] Nano architecture, if you will. And so they do that, like with heat treating, what you're doing is by heating it, you're allowing the crystals to flow basically. And then depending on what temperature you get it up to, and depending on how quickly you cool it down, affects how the grains reform. And so you're trying to give it a structure that allows it to either be hard or tough or ductile or whatever depending on what properties you're looking for. So what they did, they took titanium alloys, took multiple titanium alloys, they shocked it with these millisecond shocking, and that itself aligns the microstructure with the electrical current. That's how it works. It's basically moving the atoms of the metal into alignment at the nanoscale. And this increased the strength, well, the strength by 13 to 14 percent, the ductility by 12 to 13 percent. So no trade-off, they both improved.

Speaker 6:
[111:15] Damn, man.

Speaker 2:
[111:16] And the total energy used was less than half of a typical heat treatment for the same alloys. So yeah, pretty good. And so it's a quick, more energy-effective process that gives you, without any trade-off, you get a both increase in ductility and strength. Now these titanium alloys are used in all kinds of things, like aerospace, you know, and so that where you need light, strong metals. And so anything that could make them stronger, you know, then is, could have a lot of applications.

Speaker 6:
[111:44] Steve, how does grain size, I know grain size is also a critical component here, right? The smaller, the smaller it is, the tougher the metal is, right? Is that how it works? The smaller, the better in a lot of...

Speaker 2:
[111:55] The smaller, the harder.

Speaker 6:
[111:56] Right, the harder.

Speaker 2:
[111:57] The larger, the stronger. So, that's for steel. I don't know if that applies to titanium. But for steel, when you mentioned quenching, you want it to harden so fast that larger crystals don't have time to form. So they're all very small. And that makes it very hard, but also very brittle. Whereas if you have larger crystals, you cool it slowly, you temper the metal, then that makes it stronger but softer. Usually what you do for steel is you quench it to harden it, and then you temper it to take away some of that hardness, so it's not brittle.

Speaker 4:
[112:37] It's so interesting. I guess it's not different. It's all the same principles, but we anneal silver to make it more workable, but it's way softer. And then when you hammer it, when you work with it, it becomes work hardened and needs to be softened.

Speaker 2:
[112:52] Well, yes, some metals are hardened by hitting them because it compresses them. They're compression hardened. And that includes silver, and that also includes bronze.

Speaker 4:
[113:02] Yeah, bronze over any jewelry metal is like that.

Speaker 2:
[113:05] Yeah, if you look at like a handmade bronze item, it looks like it has all these hammer marks on it, and you think it might be an aesthetic, and it is, but it's also functional. They're compressing the bronze to make it stronger.

Speaker 4:
[113:17] Yeah, otherwise you just have to like rub something against the surface back and forth a lot, and that can strengthen it as well, but you have to do that with jewelry.

Speaker 2:
[113:24] Yeah, so you can mechanically strengthen it, or you could heat strengthen it, or electrically strengthen it, I guess.

Speaker 6:
[113:29] Or you could dope it with other metals like gold to make it less ductile, like pure 24-karat gold. You don't want any jewelry like that because it's too soft.

Speaker 4:
[113:40] Same with fine silver. 999 silver is way too soft, like you can't get it hard enough. So only the bands, those little bezels around stones are made of fine silver because they're soft enough to like push to the stone, but you'd never want the shank of your ring to be fine silver. It would deform.

Speaker 6:
[113:56] Good one, man.

Speaker 2:
[113:57] Titanium or whatever. Yeah, that's what alloys are for.

Speaker 4:
[114:00] Yeah, exactly.

Speaker 2:
[114:00] Okay. Good job, Abhin. Kara, Evan, give us a quote.

Speaker 5:
[114:04] We will explore. We will build ships. We will visit again. We will construct science outposts. We will inspire, but ultimately, we will always choose Earth. We will always choose each other. Christina Cook, Artemis II Astronaut.

Speaker 2:
[114:21] Nice.

Speaker 4:
[114:22] Yes.

Speaker 2:
[114:24] Yeah, we are explorers, but home is always home.

Speaker 4:
[114:26] Yes, there is no plan B.

Speaker 2:
[114:28] And it's more true of the Earth than anything else. I mean, the Earth is more hospitable. The most inhospitable place on Earth is more hospitable than any other place that we know of in the Universe.

Speaker 4:
[114:39] Yes, it's why our scientists at NASA study the Mariana Trench, study like the most extreme parts of Earth just to prepare for space travel. And it's still not good enough.

Speaker 3:
[114:52] Yeah, right.

Speaker 2:
[114:53] All right. Well, thank you all for joining me this week.

Speaker 3:
[114:56] You're welcome, Steve.

Speaker 2:
[114:57] And until next week, this is your Skeptics Guide to the Universe. Skeptics Guide to the Universe is produced by SGU Productions, dedicated to promoting science and critical thinking. For more information, visit us at theskepticsguide.org. Send your questions to info at theskepticsguide.org. And if you would like to support the show and all the work that we do, go to patreon.com/skepticsguide and consider becoming a patron and becoming part of the SGU community. Our listeners and supporters are what make SGU possible.