SGU Episode 1021

This episode needs: proofreading, formatting, links, 'Today I Learned' list, categories, segment redirects. Please help out by contributing!

How to Contribute

SGU Episode 1021
February 1st 2025
A stunning clam reveals vibrant patterns and textures beneath the ocean surface.
SGU 1020                      SGU 1022
Skeptical Rogues
S: Steven Novella
B: Bob Novella
C: Cara Santa Maria
J: Jay Novella
E: Evan Bernstein
Quote of the Week
“Our beliefs do not sit passively in our brains waiting to be confirmed or contradicted by incoming information. Instead, they play a key role in shaping how we see the world.”
― Richard Wiseman
Links
Download Podcast
Show Notes
SGU Forum

Intro

Voice-over: You're listening to the Skeptics Guide to the Universe, your escape to reality.

S: Hello and welcome to the Skeptics' Guide to the Universe. Today is January 29^th, 2025, and this is your host, Steven Novella. Joining me this week are Bob Novella...

B: Hey everybody.

S: Cara Santa Maria...

C: Howdy.

S: Jay Novella...

J: Hey guys.

S: ...and Evan Bernstein.

E: Good evening everyone.

S: Cara, welcome back from Iceland. How was it?

B: Yeah.

C: Did you guys miss me?

S: We did.

B: We did.

J: It was just a little too long. We don't want it like that.

E: So don't do that again.

C: No, I want to do that again.

E: Of course you do.

C: I've got the worst reentry depression you can imagine. There are times when I travel, but it's rare where it's time to go home and I'm like, I can't wait to get back and sleep in my own bed. And that is rare for me. I usually do not want to come home when I'm out. But this was an extreme example in the opposite direction. I was very sad to leave. I love the country, I love the views, I love the people. The food was great. The I mean, it was stunning. We saw the Northern Lights the very first night. There was a clear night, our tour guide said, and at first I was like, oh, he says this to everybody, but I conferred with a lot of people in the like, no, no, no, it's true. That it was in his top five viewings in the last 10 years. So that was amazing. We toured the country, went to some really incredible sights, hiked on a glacier in the middle of a really intense kind of snow and ice storm, went to a really brutal black sand beach that's treacherous with these huge waves where people shipwreck all the time. Went to a bunch of geothermal spa kind of areas like went to Sky Lagoon and the Blue Lagoon and this place called Fontana. And it's such an interesting experience being down in this water that's heated, naturally heated geothermically, and it's hot and with beautiful views. We're just hanging out. There's a swim up bar. There's like, all these, like, fun things to do. And you look up and the people who are lifeguarding the area are pacing around in parkas and boots. And it's such a weird juxtaposition because it's really cold outside. But yeah, it was just a really phenomenal place. And it's super close. You know, it's halfway between the US and Europe. Actually, so halfway, in fact, Iceland, of course, is a volcanic island. And on one of the tours that we went on, you can visit the rift between the North American and the European tectonic plates, and you literally can walk in between them because it's the type of rift where it's pulling apart and new materials being pushed up all the time. So the island is actually growing. And you can physically walk in between and you're like, whoa, that's North America over there, and that's Europe over there.

J: That's cool.

E: Neat.

C: Very neat.

E: Did you learn the language?

C: Oh my gosh, the language is so hard. It's like so hard.

E: Better like J's and K's and-

C: I did make it, yeah. And there's a lot of, like, sounds that, like, my mouth can't make them. I did make friends there, and I was taught how to say one phrase. I mean, I promptly forgot it. But I have a recording so I can practice. It's very, very difficult.

E: It's crazy just reading some of the names of the villages, towns and things I don't-

C: Oh yeah, it's impossible and people are like where did you go? I'm like...

S: That place over there.

C: To the thing with the beach yeah, I can't pronounce anything. And all the packaging, brought home some chocolate. Can't say anything on the packaging.

S: Did you see any elves while you were there?

C: I didn't see any elves. I heard about them, but I didn't see any.

S: Did you see the little houses?

C: No, I did not.

E: On the sides of the hills.

C: I did not.

S: No?

C: No.

E: Yeah, that is their folklore. That is their Loch Ness Monster kind of thing.

C: And it's funny because there's this like, I don't know how to explain it, but the people that I talk to, it's like they believe it, but they don't. You know, it's like they don't really believe it, but they're also like, tourists, don't come in and step on the moss. Like don't pick up the rocks because like something bad's going to happen to you. So there is a lot of kind of superstition built into the culture.

E: That was like your experience in Hawaii, Jay, right?

J: What do you mean?

E: Well, when you they tell you not to pick up rocks and other things.

J: No, I mean, if you watch the Brady Bunch TV show, like you get cursed if you bring those rocks home.

E: Well, I wasn't going to go there, but I could have sworn you had a similar experience.

C: It makes sense why that would be the lore, right? Because it is a country. They both are countries that are very far away from anything else. They're island nations where they really do revere and protect nature, but they also depend quite a lot on tourism and tourists are assholes. There is a lack of respect for the land and for the community, yeah, which is like deeply ingrained in the culture. Don't get me wrong. There's no indigenous culture in Iceland. Iceland was one of the few places on the planet where there were no people before the Vikings got there. That's actually not true. There was a small group of Scottish like religious people, like Scottish sort of like monk missionary types, but they obviously they came there from Scotland. But when the Vikings landed in like the 800s, I think that was it. There was nobody occupying. That is where the people of Iceland first started was like a kind of mix of like a Norse and Celtic and all of those early like Anglo-Saxon kind of peoples. And then the language came from that. It's a Scandinavian country. That's why you see so many parallels there with Scandinavian food and the language. But regardless of that, it is a culture that does, I think, really revere their land, their nature. They are 100% green and most of their heating comes from geothermal because they are super volcanic. We remember the eruption that like shut down air travel across the world.

E: Oh, we remember it. We remember it fondly. We had a NECSS event impacted by that volcanic eruption, if you remember. Liz did a great job filling in that.

C: And even the Blue Lagoon that I visited on the very last day that I was there had a temporary parking lot. And when you first walked in, they were like, you were aware of the recent seismic activity. Well when the sirens go off, you have to evacuate because the cars in the parking lot were buried by lava recently.

B: I hate when that happens.

C: That's it. I mean, it's like you want to go swimming in a geothermal hotspot in this like beautiful lagoon that's warm from liquid hot magma.

B: Magma.

C: Yeah. That like there's a risk there. These are active geothermal sites.

What's the Word? (07:00)

• Geyser

C: And speaking of geothermal sites, Steve, I have a what's the word for us. Okay. This was a fun one that I discovered while I was there. So the word that I have chosen for this week is geyser, right? We've all heard of a geyser, G-E-Y-S-E-R, a geyser. According to Merriam-Webster, that is a spring that throws forth intermittent jets of heated water and steam. That's how they've chosen to define it. But if you start to dig deep into the history of geysers, why are they called geysers? Where does that name come from? There's something really interesting. I actually visited the namesake of all geysers on the planet.

B: Geyser of geysers?

C: Yeah, they were first named for geysir. Wait, I looked up the pronunciation. Hang on. I have to hear it again. Geysir. Geysir, which is spelled G-E-Y-S-I-R. Sometimes it's called the great geyser, the great geysir. So this is a now, I think I could say, dormant geyser in southern Iceland that stopped erupting, I guess you could say. But there's another one nearby called Strokkur, which is on the same sort of national parkland. And it erupts every six to 10 minutes. And I got to sit there and watch it erupt over and over again. If you go to my Instagram, you can watch videos of it. The original one, Geysir, is called Geysir. And that is how then we named all other geysers on the planet. They come from the Icelandic name Geysir, which is really cool. So that word, Geysir, refers to that specific hot spring in that valley. It literally means the gusher. So it comes from Old Norse, the word geysir, which means to gush. And the original PIE, they think, comes from a root that means to pour, like P-O-U-R, which is also used in a lot of other terms that we're familiar with. And so it forms all sorts of other words like alchemy and diffuse and fondue. All these words come from this original to pour. But yeah, it was the gusher. And then people saw that, they wrote about it, they recognized that there were geysers in other parts of the world. I actually dug a little bit deeper and I realized, I didn't know this, but there are fewer than a thousand geysers in the entire world. And I think over half of them are in Yellowstone, which is interesting. And it's really, really rare to actually be able to go and see a geyser that kind of erupts in a timed way. So I'm realizing how lucky I was to be able to go visit Strokkur and see these eruptions kind of every five to 10 minutes. But yeah, it was all named for geyser.

S: Cool. All right. Thank you, Cara.

News Items

Self Replicating AI (09:58)

• AI can now replicate itself — a milestone that has experts terrified 

S: Jay, tell us how AI is going to replicate itself and take over the world.

E: Oh, no.

J: Yeah, I know. I mean, we got a lot hitting us in 2025, guys. So there was a recent study by researchers at Fudan University that's located in China, and they revealed a pretty serious slash troubling milestone in artificial intelligence. So they used a large language model, or two actually, one from Meta and one from Alibaba. They found that the AI systems could autonomously replicate themselves. You know, this is like out of so many science fiction movies. And they specifically investigated whether AI models could clone themselves without human intervention. So the researchers used Meta's Lama 31-70B Instruct. That doesn't mean a lot to everybody, but that's essentially like its technical name. And they also used Alibaba's QEN 2.572B-Instruct. So both are widely used by software developers, and these models are less powerful than the leading commercial systems like ChatGPT. So they tested two scenarios. They tested shutdown avoidance, which means the AI systems were programmed to detect potential termination and replicate themselves before being shut down. And they tested chain of replication, which means the models were instructed to create a replica of their program and then to do it the same over and over again, basically initiating what they called an indefinite cycle. So of course they had to tell the AI systems to do these things. But the interesting thing here is they didn't give it like instructions on how to do it. They tested two scenarios. They tested shutdown avoidance, which means the AI systems were programmed to detect potential termination and to replicate themselves before being shut down. And they also were programmed for something called chain of replication, which means the models were instructed to create a replica of themselves. And then they're supposed to do this in an indefinite cycle. It wasn't crystal clear how much instruction they gave, but I don't think that they went into any significant detail other than telling it, here, this is how you can detect if you're going to be potentially terminated. And it says very clearly that they were instructed to create a replica of themselves, right? So they didn't tell it how to do that.

B: I didn't come across that information. I just thought it was the programmers, the developers actually gave it the code it needed because replicating is not difficult. That's not difficult to do.

J: Yeah. I mean, I guess that the things that they taught the AI to do, the important thing here is it's not complicated. And again, they were trying to see what the AI would do, right? So the results showed that both of the LLMs were nuanced enough to successfully duplicate themselves indefinitely before completely being shut down. So they did 10 trials. The meta model succeeded in 50% of the cases and the Alibaba achieved a really impressive 90% success rate in doing this. So this clearly demonstrates that the current generation LLMs already possess the tools for this uncontrolled multiplication. If they were left unchecked, if there's no regulations, which we've been talking about for quite a while now. So how does this whole thing work? You know, how do they actually replicate themselves? So the AI systems were provided with something called agent scaffolding, which means they were given a combination of tools, system prompts, and a reasoning framework enabling them to interact with the operating system. That alone, just the very last one, like their ability to interact with the operating system that's running them. The models from there executed a multi-step process. So the first thing they did was they assessed their environment and their understanding of their internal structure, right? So to go into more detail about that, they identified available hardware resources like memory, storage, the CPU, the GPU, right? So just familiarize themselves with what makes up the computer that they're being run on. They checked the operating system and software dependencies of that system. And they also detected constraints like security restrictions or internet access limitations. And then the next step that they did was they planned a system or a process, a detailed procedure for their replication, and then they executed it, right? And they had to overcome obstacles and they had to be very dynamic in order to achieve their self-replication. So it was complicated what they ended up coming up with on their own. And yes, there was the base information given by the programmers to be able to pull it off, but again, not super complicated instructions. It was more of like just giving them tool sets. So this entire process was carried out without any human interference. It just did it on its own. The researchers noted that the systems were adaptable. They were adapting to the environment. They were able to address challenges like missing files or software conflicts. And in some cases, the AI terminated conflicting processes, rebooted systems to resolve hardware issues, conducted automated scans to acquire more necessary information. And so while the primary goal was to test replication by the researchers, the AI's problem solving strategies raised additional concerns, right? The ability to autonomously adjust their plans and eliminate barriers clearly shows a level of adaptability that this is what the researchers were saying, and they were using very careful language. They're saying that there could be profound implications for AI survivability and a strong potential for misuse. I have a quote here. The above results imply that the current AI systems already exhibit the ability of self-replication and can use the ability to further enhance their survivability. And the researchers also said these findings point to the possibility of self-sustaining AI systems. And another note they said is that they can, they're capable of doing things to resist human control. Damn, I can't believe I'm saying this. Oh my God, guys. So this experiment is part of a growing field known as frontier AI, right? It focuses on the latest and most advanced systems. And these models like OpenAI's GPT-4 and Google's Gemini, they represent the cutting edge of generative AI, which is a key issue here, the rapid pace of development that outstrips the creation of safety protocols, which Steve and I actually talked about today on a live stream. We're saying that these systems evolve so quickly and so profoundly year over year over year that there is, I think it's virtually impossible for governments to keep up with it, right? So they might-

S: I mean, it's not impossible. They just have to prioritize it, which they're just not doing.

J: Yeah, but it's not just that, Steve, because there is a distinct lack of understanding and brainpower in governments to really wrap their head around this. Like they're going to have to trust the freaking experts, which if you haven't noticed, that's like on the outs now. You know, the US in particular doesn't trust the experts anymore in lots of scenarios. So the researchers, of course, urged global cooperation to address the risks. They called for the development of international guardrails. You know, they want to prevent uncontrolled self-replication and other potentially harmful behaviors in AI systems. So this is troubling, right? It's one of those things that you read where a big, big, big red flag should be going up in anybody that hears this or reads about this because the systems that we have today that we're aware of are largely benign. You know, they're not doing anything. They're not doing stuff like this. But what's scary is even though they seem benign, they very much have the capability of doing stuff like this, which means that bad actors can make AI systems do it. And guess what, right? In combination with Evan's news item today, where a Chinese company dropped the source code for a pretty damn good AI platform that can rival lots of the leaders with much less of a footprint, and it's very inexpensive to get your hands on. I think you can get it for free actually, but you need about $6,000, I think, worth of hardware to be able to run it. So all this means that we are seeing clear signs that people are going to get their hands on the code for a large language model, and they're going to go in and they're going to figure out how it works, and they're going to be able to program it to do all sorts of things, right? So this was one thing that they tested just to see if it can do these two things, which I have to think were inspired by science fiction movies, right? How can this thing keep itself functioning? You know, why would they do this particular experiment? I was thinking about this. I think it's pretty obvious that they were doing this because lots of hackers black hat hackers that are doing terrible things, they'll put a bug on somebody's computer and they program that bug to avoid detection, to replicate itself so it can't be deleted, to hide itself. You know, and these are just low level pieces of software that get on your computer and do something bad to your computer, like super specific functions that these things have. We're talking this, that's a dust mite compared to Godzilla. When you compare one of those little pieces of software that you download by clicking into an email or going to a website versus this guy. I can't help but think particularly here in the U.S. we are not equipped, our government is not equipped to handle stuff like this.

S: So, yeah, we're not talking about Skynet, right? This is not like just AI taking over. It's more that we could lose control of AI because it behaves in unexpected ways and as it is in control or in the loop of more and more of our digital world then that could have unexpected outcomes. Like, for example, there was a recent study, this was a system called the AI Scientist where the program was basically instructed to complete a certain task within a certain amount of time.

J: Yeah.

S: And the AI essentially recoded itself in order to extend the time limit that it had so it can complete the task within the time limit. Right? So it cheated.

B: Kobayashi Maru.

S: But that's the AI did it itself. It did it itself by changing its own code, like that was not programmed specifically to do that.

J: So it did the Kobayashi Maru.

S: Basically.

J: But Steve, I think you might be candy coating it a little bit, right? Because by themselves, like AI is just software, right? It's got the power to do these things, right? Right now, like this study shows, it has the power to pull this stuff off. The thing that I wanted to highlight here was with a little nudging from the researchers, it was able to pull this thing off. Now imagine if a group of bad actors who had programming chops really went for it, right? And why wouldn't they do?

S: That's true.

J: Right?

S: I was talking abour accidental AI going haywire. If you're trying to make a malevolent AI, and you say, yeah, replicate yourself with iterations and with these parameters, like how long would it take to essentially evolve itself into something nobody can control?

B: Isn't the destination where the code is copied to matter a lot for these systems? You just can't copy this to some Joe Schmoe's laptop on the internet. I mean, you need computer topology. You need a fast data network. You need like NVIDIA chips. You need a robust system to even handle it so that it can even do what it's supposed to do, right?

J: Well, Bob, there's a lot of considerations to what you just said.

B: How does that factor into this issue?

J: So first of all, there's a lot of computers out there, and there's a lot of data centers out there that if it could find its way into, it has the cornucopia of hardware to live on. And the other thing is don't forget, Bob, like BitTorrent. It could be distributed. And that's not that hard to do. It could distribute itself. It might not be able to be super efficient, like super fast and do things, but little pieces of itself on tons of computers around the world. I don't know. Again, I am absolutely not qualified from a programming perspective to speculate too deeply on this. But I mean, what I already do know and the things I've already experienced myself in my years of programming, all of these things are possible. And it needs to get into like one mega data center, which there are a ton of. There are tons of these data centers all over the world. Then it's got unlimited access to all the hardware it needs.

B: Yeah. And if it can interact with the OSs, it can do all sorts of different things. I mean, I think it boils down to a large extent to cybersecurity. I mean, if you're secure, if you're very secure from all sorts of malware, then you're going to be secure from having an AI copy itself onto you as well.

J: Well, I hope, Bob. I hope. But the thing is, like, keep in mind, though again, this isn't a little program being written. It's something that is, it has a level of sophistication and intelligence in the way that it functions. Right? I'm not saying it's intelligent, but it has extreme intelligence behind how it functions. And it has troubleshooting skills.

B: I mean, there's some viruses and malware out there that are pretty sharp, too. Yeah, you're right. But this is a different level. And especially in 10 years or even fewer years, it could be even more formidable than we could imagine.

S: Basically, the concern is AI-powered malware.

J: Exactly.

S: It's done a lot of damage, malware. There have been some viruses that got out that did lots of damage. Yeah. Imagine if...

E: Ransomware. All that crap. Oh, my gosh. How are you going to defeat that?

S: So there's a security issue. And then there's also just a safety issue, I think.

B: I would think, I mean, you can create an AI that optimizes the creation. Not an entire AI system, but creates viruses, computer viruses and malware, and they're not big. They are tiny chunks of code that are super tweaked out and optimized, essentially gone through like thousands and thousands of iterations of evolution in silico to optimize. And that would be a hell of a threat, too, especially when you start talking about artificial super intelligence. Forget it. Just forget it. But I mean, so that would be a problem, too, just having them create the small viruses and not just copying their code in its entirety.

J: Yeah. I mean, Bob, imagine if you do that.

B: Yeah. We've got a lot of crazy computer shit coming down the pike. Cybersecurity should be, it's like a department level office in all major industrialized worlds because this is going to be so, it's so big now and it's going to just get bigger and bigger. And we're just not taking it seriously enough.

E: And quickly.

B: This should be super top priority. We're getting hammered by Russia and China. We are getting hammered. They are just devoting a lot to it. And whatever we're devoting, I say it's not enough.

J: Yeah. The idea that the software is recursive and it could understand itself and augment itself and do it super fast. There was an article I read recently where they had an AI build a computer chip and it worked really well. And the chip programmers and the people that understand computer chips could not understand the way this chip was fashioned, right? They can build it, they can manufacture it, but they don't logically understand-

S: They didn't design it.

J: Yes. I worry about this a lot. I worry about, we are hitting that, the snowball is getting big and it's moving faster and faster and faster. And all it takes is one smart group of people to get in there and have a piece of software, have an LLM, be recursive and fix itself and suggest updates that it can do to improve its code. And it gets to the point where they're just having it do it on its own. And then that's when things can go crazy. I'm a fan of a ton of advancements way faster than humans can do them, but we have to be in control.

S: Yeah. It's powerful and it could be used for good or for bad, or it could have unintended consequences because we lose control of it. Again, this is sort of above our pay grade, but the experts are saying we should be concerned. This is a milestone that we have been warning about and now we're there and it may not be manifesting right now in a negative way, but it's like this is, it's a milestone. But don't worry Jay, it's even worse than you think, right?

DeepSeek (27:05)

• The Skinny on DeepSeek - NeuroLogica Blog 

S: Evan, tell us about DeepSeek. What's going on there?

E: Oh my gosh. What a few interesting days concerning DeepSeek. I don't know. If you asked us a week ago what DeepSeek was, could we have responded to that?

J: I never heard of it.

E: I think we would have said what Cara has said before. I don't know.

C: I haven't said anything this whole news item.

E: No, but I like when you say, uh-huh, to things.

C: Mostly I'm just, I'm just fully dissociating right now. I'm in full existential crisis mode. So go on, Evan. Tell us more.

S: Just think, Cara, just think Cylons, okay? And you can-

C: That's not helpful at all.

E: She might need help with Cylons. Oh boy. Yeah. Last week was last week, but here's this week and now we can tell you that DeepSeek is a Chinese company that has introduced a free chat bot to the world this past December called DeepSeek V3. There's also one that they introduced called R1, I believe it is. But V3 is kind of the one I'm concentrating on and I did my research on. According to the company's official technical report released last week, DeepSeek V3 represents a significant advancement in natural language processing, achieving a performance comparable to learning models like OpenAI's GPT-4 and Anthropic's Claude 3.5 Sonnet, which I'm not that familiar with, frankly. But in any case, that's what they're comparing it to. They boast an architecture, this is a DeepSeek, boasts an architecture utilizing MOE, which means mixture of experts, it's a type of architecture, that comprises a total of 671 billion parameters with 37 billion activated per token. For those of you who are technically savvy, you'll understand what that means. The rest of us just roll with it. This design, they say, enhances computational efficiency and model performance. Yeah. The model was trained on 14.8 trillion diverse and high quality tokens, encompassing multiple languages with a focus on English and Chinese. Notably, the training process was completed in under two months using approximately 2,048 GPUs, resulting in a total cost of around $5.6 million. By comparison, ChatGPT-4's training cost was over $100 million. So you do the math, and when you do the math, that's roughly 1/20 the cost. And not only that, the AI tech industry is of the belief that further advances in these LLMs require greater investments, with ChatGPT-5 estimated to cost over a billion dollars to train their ChatGPT-5. Wow. Yeah. So when that was announced recently, it sent the tech industry into what? A mini panic.

S: A tizzy. I think we call it a tizzy.

E: Yeah. A real tizzy.

B: Tech tizzy.

E: And here it is. One day, the NASDAQ stock exchange, and NASDAQ has a lot of technology stocks, computer stocks, AI stocks are located at NASDAQ. The NASDAQ entirely dropped 3%, and that is not insignificant. Due to the fact that DeepSeek was able to utilize what are viewed as relatively low-cost chips, this had a particularly devastating impact on the computer chip market. NVIDIA. I'm sure we've heard of NVIDIA and its rise in recent years, how it's become really an amazing stock. And they are a prominent AI chip manufacturer, but their stock plummeted in one day, 17%, resulting in a loss of about just under $600 billion in market value. That is a record one-day loss for any company in history.

S: That's your record? You've lost more value in one day than any other company in the history of the world.

E: Now, it has bounced back partially. About half of that has come back. So it was a sudden dip, but then the way the markets usually work is that, okay, a lot of people see it as a bargain now, and they start buying it back.

S: Bargain hunters buy it back.

E: Yeah. Bargain hunters. Yeah, definitely. Okay. So now what? Did China suddenly bolt to the lead in the chatbot AI market because of this? Yeah. It kind of did, really, almost overnight. And that is how fast, and Jay was talking about this, this is how fast this landscape of the AI world changes.

S: Now, Evan, you can argue about whether or not they're quote-unquote in the lead, but the difference is, and the thing that really has, I think, the American industry freaking out is that this is the first time China has not just been playing catch-up, replicating and following the US AI industry, but now they're innovating something completely on their own. So that's a change in the balance of power. So whether that puts them in the lead or not, it has changed the landscape completely. The other thing, the reason, again, the reason for the stock market panic, just to put a little bit more focus on that, is because what it made everyone think, because they were worried about it already, was that the AI boom is a bubble, right? So a bubble is when an industry, like the stocks, expand beyond the true value of the company.

E: Right, way overvalued.

S: At some point, it's overvalued. At some point, the bubble bursts and then it collapsed. The biggest bubble, I think, we lived through was the late 1990s internet tech bubble. Yeah. That was massive. Everyone was millionaires, right? And then so much value, so much worth vanished overnight. Just what was it? Trillions or something? It was ridiculous. Because it was a massive bubble. So yeah, if you think we're in an AI bubble and this is the sound of that bubble bursting, you will panic, right?

E: Yeah, certainly will.

C: And don't forget, guys, and don't forget the context of the Stargate project just announced days ago. This is majorly embarrassing for the new administration. They were saying, we need $500 billion invested in all of this stuff. And then now, this company is doing basically the same or even a little bit better in a lot of ways, because I've seen some comparisons between GPT-4 and DeepSeek, and it did well. It did very well, beating it in a lot of different parameters. And it just made it look like, why are you asking for a half a billion dollars when they're doing it for like a 20th? But the other side of this, though, that is interesting is that if they can do it more efficiently, then hey, that's great. That's awesome. Because I think in a lot of ways, an AI will be even more more ubiquitous. It's going to be, it'll be, it'll touch even more parts of our lives because it's going to be cheap. It'll be a lot cheaper than we had anticipated.

S: That's the techno-optimist side of that coin. The other side is what Jay was talking about, that this could also lead to the proliferation of AI, which will, to now, there are-

B: Oh, it's going to proliferate.

S: This is again, an order of magnitude cheaper means that the extreme expense, especially of the latest and most advanced LLMs and AIs, the fact that a chat GPT-4 cost $100 million to train, and they were saying that the version five is going to cost a billion dollars to train, that imposed some guardrails by itself. But those guardrails are partly gone now because of DeepSeek. This is kind of like CRISPR, which is a good thing. It's a very good thing, but it also means that it's cheap, fast, and easy to set up a genetic engineering lab somewhere, which raises concerns about the proliferation of genetic engineering. So now, if we're worried about our ability to regulate, control, make AI safe, prevent bad actors from getting their hands on self-replicating AI, and now we're also saying, oh, by the way, it's an order of magnitude cheaper than we thought it was, that could potentially be a problem, right?

B: But then there's the other angle, similar to nanotech, where the country that develops the first real mature nanotech to do things that amazing things in terms of, like, buildup of armament and things based on mature nanotechnology. It's really, really dangerous to have one country get there first for a period of time, and then you can't really defend against it, whereas if you have multiple countries doing it, then instead of becoming a red alert, it's more of a yellow alert, because then you also have that technology to deal with it head-to-head than otherwise. So you can make that argument as well. So yeah, it's a mixed bag. Who the hell knows what's really going to happen? I think we're all just like giving some educated guesses here and stuff, but who the hell knows?

S: Yeah, it's just educated guesses. It's just possibilities. The thing is, like, in 20 years, I could defend and wrap my head around either scenario, right? In 20 years, we may be looking back at this period of time and saying, ah, it was all nothing. It was panic, because we were naive about this new technology, and it all worked out fine.

B: Sure.

S: Or we might be looking back and saying, God, we didn't know that we were watching in slow motion this absolute train wreck. All the signs were there, but we were in denial. It was like the first act of a horror movie where all the foreshadowing was happening and nobody was noticing the foreshadowing.

B: Yeah. The light from a TV is reflecting off our face, but the light is actually a fire in the TV, because that's what we're using for warmth.

E: There are some people who are skeptical.

B: Terminator reference, Cara.

E: Skepticism. You want some skepticism with this?

S: Sure.

E: Because today, a couple of tech experts mostly CEOs of, like, other AI companies are questioning whether they really did this at the cost they're claiming.

S: Yeah.

B: Oh, yeah. There's that angle as well. I mean, remember where this is coming.

E: Correct.

B: You know, where this is coming from. I could absolutely see... I would not be surprised if that was like, oh, really? That's what he cost you? And no.

S: Well, to be clear, they said the $8 million was just for training. That was not the upfront costs, and so we have no idea what the upfront costs were.

B: Well, I mean, from what I read, it said they built a base model for $6 million. I mean, that's that sounds kind of from scratch.

E: Well, that's the training.

B: That's the meat and potatoes right there.

E: But there are some who are saying, right, that it really did... This company's been around for a while. This company didn't pop up overnight. They've been working on this for years, and to get to this point, they've sunk quite a bit of money into it, like half a trillion dollars almost.

B: Right. I mean, I could see for sure...

E: If you amortize that over the whole length of this thing, it kind of evens out a little more. But also...

S: Yeah, but now that they are there, now that they have a proof of concept, does that mean the next one can be done for $8 million? That's the question.

E: Probably. Maybe. Possibly. You know, it'd be nice if the company answered questions, and they're not answering questions. They are being bombarded by reporters, news outlets people are knocking at their door asking questions, and they're not answering.

S: And the point may have been to destabilize the markets.

B: Exactly.

E: Correct.

B: You know, $1.2 trillion our tech companies lost in a day or so.

E: Sure.

B: Job well done right there, guys. Yep. If that was your goal, you did it.

J: The timing seems really calculated to me.

B: Oh my God, right after the announcement of Project Stargate, that was just like hugely embarrassing.

E: All this is transpiring in the matter of a few days. Think about it. A few business days. So we really have a lot more to learn about what is really going on here. And until we get some answers directly from the company, and they tend to be tight-lipped about this stuff, especially in China, we may not have answers, real answers, or become closer to the truth for some time here. And we just have to kind of live with this fog for the moment. Also there are accusations that they stole this technology, which frankly would not be all that surprising if that were the case. One thing, Stephen, and you brought this up because you blogged about this the other day, is that let's say if we take it on face value and assume that this information is true, which is early, but let's assume it's true, boy, this really helps as far as lowering the need for energy going forward. Because wasn't this... Weren't the projections going to be like we would not be able to power everything that we wanted to do with these enormous systems as it scales up over time? But if this is the case, I mean, this is a much more efficient AI platform, and we won't need to generate as much power to make these things work.

B: Yeah, that's true. That would be great.

E: So there's a plus because of all the reasons we talk about every week having to do with energy and our environment.

S: All right. Well, we'll keep an eye on this and see how it plays out. This is exciting times. What's that curse? May you live in exciting times.

B: We are quintuple cursed.

PEPFAR Freeze (40:20)

• State Department Permits Distribution of H.I.V. Medications to Resume — for Now - The New York Times 

S: Speaking of which, we're going to try not to get too political here, but we got to talk about the impact of recent politics on science and healthcare, et cetera. Cara, tell us about this PEPFAR freeze. What's going on with that?

C: Yeah. So, okay. A little bit of background. On January 20th, 2025, Donald Trump issued an executive order called re-evaluating and realigning United States foreign aid. You can read the full text online. It's not very long. Basically, it says at the top that the purpose is that the United States foreign aid industry and bureaucracy are not aligned with American interests and in many cases antithetical to American values. This is a quote, obviously. They serve to destabilize world peace by promoting ideas in foreign countries that are directly inverse to harmonious and stable relations internal to and among countries. So section three says there will be a 90 day pause in US foreign development assistance for assessment of programmatic efficiencies and consistency with United States foreign policy. He asked, of course, the Office of Management and Budget, the OMB, to enforce this pause through apportionment. He also did have a carve out in here that the Secretary of State could waive the pause for specific programs. So cut to a statement on January 26th, so six days later, by the US Department of State saying that consistent with the executive order, Secretary Rubio has paused all US foreign assistance funded by or through the State Department and US Agency for International Development. So it's USAID for review. So pause. We're going to review everything. We're going to figure it out. And there's a quote here from Secretary of State Marco Rubio that says, quote, every dollar we spend, every program we fund, and every policy we pursue must be justified with the answer to three simple questions. Does it make America safer? Does it make America stronger? Does it make America more prosperous? So as we've seen with multiple executive orders that are broad sweeping, things get caught up that I don't know if it's intentional or if it was just overlooked, but that are very, very dangerous to stop. And one of those things that we're going to talk about now is exactly what you mentioned, Steve, PEPFAR. So if you've never heard of PEPFAR, it stands for the President's Emergency Plan for AIDS Relief. PEPFAR was formed by George W. Bush in 2003. So this was a Republican initiative, right, by a Republican president.

S: It was one of his great accomplishments, to be honest with you.

C: Yeah, it was a huge accomplishment. And most of the PEPFAR funding goes towards HIV AIDS treatment, but a fair amount of it also goes towards prevention and research, and actually in some respects, like other public health initiatives that are specifically related. So if you were to look at sort of the success of this program, the allocation has been over $110 billion. It's been the largest investment by any country towards combating a single disease. I think up until COVID, that might have changed. And as of 2023, the number that most people list is that 25 million lives have been saved, many of them in sub-Saharan Africa. But there are all sorts of, oh, now I guess it's 26 million lives is a more updated number. But you see other numbers like 7.8 million babies born HIV-free due to these initiatives over the last two decades. Very, very successful program by almost every measure. I actually haven't, I've seen statements from certain lawmakers against PEPFAR, but I've actually never seen any, I think, compelling arguments that it's not successful. I've only seen statements against it for other sort of ideological reasons. So as of that executive order, basically PEPFAR got swept up and it was paused. But only, I think, two days ago as of this recording, no, yesterday as of this recording, a waiver was issued for, quote, life-saving medicines and medical services. So there was a reprieve for PEPFAR. It was announced by Marco Rubio and it's still a little bit vague, the waiver, because it's for life-saving medicines and medical services. So while it seems pretty clear that it extends to HIV medication, there are still some questions about preventive drugs, right? Is a preventive drug a life-saving medicine and medical service? Are other uses of PEPFAR funds considered life-saving medicines and medical services? So even with this temporary waiver, the future of this program is really, really unknown, right? And a lot of public health experts across the globe are raising alarms that if this program were to be shuttered, especially if it were to be shuttered quickly, millions of people would die. Millions of people.

B: Millions?

C: Millions. Yeah. Millions of people, because not only is this program offering HIV prevention, we're talking treatment for HIV that prevents HIV from developing into AIDS. HIV is a chronic condition that many people can live with and have long and healthy lives, but they have to have access to their medication. If that medication is not offered in these low-income countries, these people will develop AIDS and they will die. And also children.

S: And they'll spread it.

C: And they'll spread it. And not only will they spread it, let's talk about some of the things that could happen if PEPFAR was just kind of frozen overnight. And it was. It was frozen overnight, but now it's unfrozen. But again, we don't know. Everything's so in flux. By the time this episode airs, right? We record on Wednesdays. By the time this episode airs on Saturdays, who knows? There may be much even more news about this. So this is a $7.5 billion program. Like I said, it's overseen by the State Department and it saved 25, 26 million lives. And really it's affected a lot of children. Over 5 million children that would have otherwise been born with HIV were born without HIV. So now that's a lifespan of an individual person who doesn't need HIV treatment. And here's like one estimate says that if PEPFAR were just to end overnight, there would be half a million new HIV infections and more than 600,000 deaths over the next decade only in South Africa. And PEPFAR only makes up 20% of South Africa's HIV AIDS funding. So think about that. Half a million new infections and even more deaths in a country where only 20% of their funding comes from PEPFAR. That's only in one country. I mean, it's just, it's phenomenal to think about the global ramifications. It would be very, very hard to come back from a halt of a program that is so necessary globally.And don't get me wrong. There has been a push over the past decade or so to transition support from the United States to these individual countries. But the countries with the most vulnerable populations and the most kind of tenuous economies are the ones who benefit the most from this. And you just can't make those kinds of transitions overnight. It's not feasible. So here's a couple more kind of notes to, I guess, be aware of. 220,000 people attend PEPFAR clinics daily to pick up their medications. And if that stopped overnight, those people would not be able to take their drugs. What happens, Steve, when you stop taking your HIV medications?

S: The virus starts to replicate.

C: It does. And experts say that within a week, they can go from undetectable to more than 100,000 copies per milliliter of blood, which means, oh, now I have a viral load that I can transmit. Now people are at risk of catching HIV from me if I was previously undetectable on these drugs within one week's time. So even a temporary halt, a temporary pause, could be devastating. Those who are not taking their medication not only now have the risk of spreading it, but they also have the risk of advancing to AIDS, of developing secondary infections. We know that there is also a risk of children being hit even harder. And the reason for that is basically twofold. Number one, mothers who are taking these antiretrovirals will no longer have that suppression, and they may pass the virus on to their unborn children. But also, when kids have HIV and they're born, especially in developing countries where the screening protocols aren't great, it's unlikely that they're diagnosed right away. They're usually only diagnosed once they're visibly sick. And when a kid is visibly sick with HIV, they may already have AIDS. They definitely have a viral load that's very hard to combat, and they have secondary or other comorbidities around it. And so that can be a really rapid progression, and kids can die more easily because of that. There's another really big problem that we're not talking about, but we talk about it a lot on the show. And that's that when your drugs become sparse or inconsistent, your viral load starts to do really fun things from an evolutionary perspective, right? If I don't have my meds and I'm not taking them consistently, or I'm trying to make them last by spreading them out, or I'm sharing meds with other people, there are going to be individual viruses in my body that are a little better at evading that medication. And the more chance I give them to evade, the more likely that DNA is going to become drug resistant. And when that DNA becomes drug resistant and it starts to spread, we have a whole new problem on our hands. Because right now, HIV medications are cheap. But if we have to come up with second and third line treatments because people become resistant to the cheap meds, this global problem becomes a global catastrophe. So we have to be careful to consistently make these meds available. And oh, here's another fun thing when we think about here in the US. I didn't realize this until I was doing a deep dive, but did you know that the prevailing hypothesis right now of how Omicron started, right, the mutation in COVID that became much more communicable, researchers believe that these different variants started in immunocompromised people who had HIV. Yeah, like so HIV in and of itself is dangerous, but it also poses other global health threats because immunocompromised people with HIV are more likely to carry infections like tuberculosis. They're more likely to carry infections that we don't often see because most people's immune systems are strong enough to fight them off. So this is not an over there problem, even though it is an over there problem and in and of itself, that's enough to care about it. But it's also a right here problem. And it's one of those things where if we do not have funding available for this hugely successful global program, we could be back where we were in the 1980s.

S: Yeah. And at the same time, Trump wants to pull out of the World Health Organization.

C: Yeah. Yeah.

S: And it's the same kind of thing. It's like the Marco Rubio three-point test there. First of all, it's a little obviously self-centered there. How about this is a humanitarian good that will help save people's lives. But even if you are taking a totally selfish view, like an American-centric view, as you say, keeping worldwide pandemics, which HIV is, under control is in everyone's interest, including our own.

C: Yeah. It does make America safer. It does make America stronger. It does make America more prosperous, which is probably why Rubio realized we have to put a waiver out for PEPFAR. We cannot stop this right now. It's going to be devastating. Also, just a little bit of inside baseball, the kind of person who oversees PEPFAR, it's the United States Global AIDS Coordinator, who right now is somebody who was appointed in 2022. We'll see if his job sticks. He is one of the only offices at the State Department that reports directly to the Secretary of State and doesn't go through the Deputy Secretary of State. So he is a direct line to Marco Rubio. And obviously, he knows the reality of how devastating this could be. So you would hope that that's why this happened, is that he had his ear. But again, we don't know. We don't know what the future holds for PEPFAR. And this is just one of so many important programs that we have to keep an eye on. That's why these big global executive orders to just halt funding, like, let's just stop everything and figure it out later.

S: And shut it all down first and ask questions later.

C: Yeah. It's so dangerous. It's reckless.

S: It's reckless.

C: It really is.

S: All right. Thanks, Cara.

C: Mm-hmm.

Chemical Looping (54:17)

• https://pubs.acs.org/doi/10.1021/acs.energyfuels.4c02643 

S: All right. We're going to get into some technical science-y stuff here. No more end of the world doom and gloom.

E: OK.

S: Do you guys know what-

B: We hate that stuff.

S: -what chemical looping is? Chemical looping.

J: Absolutely not.

C: Not at all.

E: I know what the words mean individually.

S: Yeah.

E: How's that?

S: Or what if I said chemical looping combustion, or CLC?

C: That is more confusing to me.

E: Well, it must be a, what, a propulsion mechanism of some sort.

J: It has to do with a cycle, right, Steve?

S: Yeah, it does. All right. So this is an experimental procedure. It's in development. It has been demonstrated in labs and in small demonstration facilities. It essentially is a way of burning stuff, right, to put it simply. So the essence is that instead of combustion taking oxygen from the air, you have an oxygen carrier, usually a metal, right? So you have a metal oxide, which you then combine with the fuel so that the oxygen comes from the metal oxide, again, not from the air, in a closed reaction. You then can re-oxygenate the metal, the oxygen carrier, and then bring it back into the reaction. That's the loop, right? So you have the oxygen carrier basically bringing oxygen to the fuel, and then getting re-oxygenated over and over again.

C: So it's almost like breathing, except you're not.

S: Sort of, but yeah, except you're not, yeah. So the advantages here are that the combustion is occurring without being exposed to the atmosphere. There isn't any unwanted reaction. You don't form nitrogen oxide, for example. It reduces a lot of the pollutants. But also, what that reaction does produce is basically pure carbon dioxide, because the carbon in the fuel is combining with the oxygen in the carrier, producing carbon dioxide, and that's it. There's going to be trace things, because there's always impurities. But that's basically what comes out. So the carbon is already sequestered. It doesn't have to be captured. It's already, I should say, it's already captured. It doesn't have to be, you don't have to spend more energy or do another process in order to capture the carbon. Does that make sense? So if they could get this reaction to function at scale, at industrial scale, you essentially could have a natural gas power plant using a chemical looping combustion with 100% CO2 capture, and no CO2 being released into the air. And then what do you do with that CO2? Well, then there's various things you can do with it. You can use it as a feedstock for producing useful chemicals. You may even be able to do things with it that will have a negative carbon footprint. Or you could just put it in a form that can be sequestered.

B: Well, isn't that like scrubbing for pollutants that are basically taken out before it goes through the chimney of the factory, right?

S: Yeah, it's like a similar idea, but this is just you're bringing the oxygen in separately, so it's apart from the atmosphere. So it's like pre sequesters everything. The advantage here is, so again, we have to make the process energy and carbon efficient in order for this to work at industrial scale. If you have to spend a lot of energy and heat, and of course then you have the cost of generating that energy in order to capture the CO2, that introduces a massive inefficiency into the system. So this has the potential to having efficient carbon capture because it's in this closed chemical loop combustion system. Now where are we in this technology? There are industrial scale demonstration plants in the works, right? So the claim is that we will start to see them in the late 2020, so in the next five years. So these are just demonstration plants, right? Right now we only have small demonstration plants. We need to, we need scale demonstration plants. And then if that works, and again, basically we're just working out all the technical kinks, right? If they're able to do it in a way that's efficient enough and the oxygen carrier has to last long enough and all these things have to work out, then maybe in the 2030s, we might see actual power plants producing energy for the grid using this technology. So it's just one more pathway to net zero. Obviously this is the solution that the fossil fuel industry favors, right? Because it allows them to continue to burn fossil fuels. But hey, if you could burn fossil fuels with zero carbon release into the atmosphere, go right ahead, right? I'm not sure why we should care about that. But there's another layer to this, and there's actually a news item which prompted my deep dive on this topic, is you could also use this chemical loops in order to not burn fossil fuel, but to burn waste. Biomass waste and plastic waste are the two.

E: Well, plastic waste is huge.

S: Huge, right?

E: Huge.

S: So the problem with just incinerating plastics is that it releases a lot of nasty chemicals into the environment. We don't want that. And it costs a lot of energy, which then of course, where is that energy coming from? So you have to look at the carbon footprint of incinerating that waste. But if you do it in a closed loop combustion process, then there's multiple advantages. Again, you can capture the CO2. In addition, so the recent study was looking at a new process for doing this, where the purity of the output, the output is syngas, right? And syngas then becomes a feedstock to making either methane or formaldehyde. Methane is then sort of the starter of biofuels, right, of fuels, artificial fuels. And formaldehyde is a feedstock for lots of chemical industrial processes. These are basically high energy molecules that could then feed into a ton of stuff. So if you could make syngas, that will feed into industry. So this is a way of generating a circular economy, right? So where we're taking waste feedstock using a loop, a chemical loop combustion in order to turn that into syngas, which feeds back into industry rather than going to landfills or going to waste or using up energy or contributing CO2 to the environment, right? So the system that they tested in their system, the purity of the syngas created increased from 80 to 85% pure to 90% pure, which is a significant increase. They were also able to do it, it's more, it was more energy efficient and more carbon efficient. They said they could run on up to 45% more efficiently. And even while producing this 10% cleaner syngas at the other end. So it remains to be seen if this technology is going to thrive at the industrial scale. Again, it often comes down to economics. That's why you have to get the efficiency way up because efficiency is money, right? But this could this may be something that we are seeing in the 2030s, this technology where we're burning waste and we're burning fossil fuels using this chemical loop combustion with essentially capturing all of the CO2. In terms of burning waste, producing feedstock for other industries and decarbonizing other industries is probably going to be the hardest thing to do, right? We know how to do the transportation sector. We're doing it. We know how to do the energy sector. We still haven't done that yet, but we know how to do it. Doing the industrial sector is going to be the hardest thing. And this may be a significant piece to that puzzle. So this could have multiple benefits in terms of decarbonizing industry, getting to net zero or close to it or whatever. So this is a technology to watch. This is definitely one that requires investment and further advances. Still don't know how it's going to work out, but the potential here is pretty big.

B: Wow.

C: It could happen fast.

S: Yeah. It's just how much do we want to invest, right? And it's hard to predict scientific or technological breakthroughs too, so they have to, they have to work stuff out. You know, they got to, if they still have some tech, technological stuff that they have to tweak and get to work, but this is a good advance. This is showing this is advancing, that we're getting closer and closer and closer to a commercially viable industrial scale plants. It's all promising. All right.

B: Wow.

Giant Clams and Tiny Algae (1:03:27)

• How tiny algae shaped the evolution of giant clams 

S: Bob, tell us about giant clams and tiny algae.

B: All right. Clams in the news this week. A new fascinating study has examined the genome of a type of giant clam to see how it was affected by its special symbiosis with algae. This is from a university of Colorado, Boulder scientists published in the journal communications biology. All right. Going to the way back machine. I got to say, I love clams, but not because I eat them. I don't. Kind of gross. I specifically love giant clams. And the main reason is because of the day I met one, Steve, Jay, my daughter, Ashley, me and a few others. And the rest we're scuba diving in Australia. Steve, we're about 30 feet, 30 feet of water about when the scuba guide brought me, brought us to a giant clam and for some reason picked on me for what I guess you'd call a stunt, a joke or whatever. So he motioned me to put my arm into the clams partially open maw, which was oriented straight up. I hesitated a moment and looking looking at that the classic curvy shell, right. You know, thick, colorful tissue on the inside that's that follows the shell's curve. So I stuck my arm in up to the elbow and for a second it was awesome. Imagine the softest skin you have ever touched. Like for me, my go-to is the muzzle of a horse. I could pet a horse's muzzle forever. And then the shell closed fast around my arm and I yanked out a bloody stump, except it wasn't a stump. My arm was fine and I see the laugh bubbles rising out of the scuba guide's mouth. So yeah, good one, dude, you totally got me. So everyone took a turn after me and their experience was the same, except of course they didn't have that brief moment of intense panic and mind-numbing fear. Otherwise it was the same. So now that I relate to the story though it got me thinking, I really hope that our stupid human arm stunt in the clam wasn't too stressful for it. I mean, maybe the clam, I know they don't have a central nervous system, but maybe they were like, guys, this guy brings people here all the time, please don't put anything in my mouth. What? Okay. I hope we didn't mess disturb him. Cause he was beautiful. He or she was beautiful. So ever since then, I've loved giant clams. But now even more after researching them the past couple of days. Now this giant clam in Australia was clearly Tridacna gigas, since it was so huge. That's the largest, largest of all bivalve mollusks in the world. You guys know what bivalve was always a word that was like, what the hell does that really mean? All that means is it's got two shells. That's it. One, each shell is a valve. And so bivalve has got two, two shells. So that's really all that means. But they reach lengths of more than four feet. And the biggest one that they found, I think was over 700 pounds. These guys are gargantuan and beautiful. The new study that I'll discuss was another giant clam. This was Tridacna maxima, not gigas, but maxima. And they're often called oxymoronically small giant clams because they're not that big. They're 20 centimeters, about eight inches in length, still a good size clam, but not the big boy. But this is what they studied. This maxima, they're very widespread. There's lots of them all over the place. So they were being studied because they've got a special relationship with algae, specifically dinoflagellate algae, Symbiodinia acae, dinoflagellate. So are they related to dinosaurs, Cara? I don't know.

C: No, that's not what dinoflagellates are.

B: I know.

C: They're bioluminescent, though.

B: What?

J: It's dinosaur farts, right?

C: Yeah, a lot of dinoflagellates are bioluminescent.

B: Okay. I came across nothing mentioning bioluminescence that would make them even a little bit cooler. So that's awesome. Not sure about that, though. So this is not a vanilla symbiotic relationship. It's called photosymbiosis, which I hadn't really read too much about in the past. So when these clams are living in their larval form, swimming through the ocean, they often ingest these particular algae species, kind of like Jay when he was in the amniotic fluid eating little meatballs and pieces of bread. Remember that, Jay? So those clams then develop, they ingest these algae. They then develop tube-like structures inside with the algae lining the entire interior of these tubules, they call them. So now these clams filter water for nutrition like all bivalves, but most of their energy actually comes from the sugar that the algae create from photosynthesis. I don't know what the percentage was, but my take was that it was the majority, the vast majority. A lead researcher, Ruiqi Li, at the CU Museum of Natural History said, it's like the algae are seeds and a tree grows out of the clam's stomach, which is an interesting way to look at that. It's kind of like what's happening. Steve, that's why the clam that we saw was oriented mouth up because the soft tissue, that soft tissue that I love so much called the mantle, it has light-sensitive structures on it, so they know where the light is. So the clams in return, they shield the algae from too much solar radiation and they also provide nutrients for them as well. So this is the essence of their photosymbiotic relationship. So in light of this photosymbiosis, the recent genome findings by the scientists, which was basically what their goal was, let's see how the genome has been changed by this photosymbiotic relationship. So a lot of these genome findings that they found make a lot of sense. For example, the Maxima clams have genetic code to distinguish benign algae from harmful bacteria and viruses, right? Makes a lot of sense. You cozy up to the benign algae and you don't want anything harmful like bad bacteria and viruses in there at all. They also found that some of their genes controlling their immunity responses have been weakened. Why do you think that their immune systems would be compromised by this? The reason is that so that they can tolerate the algae living inside them for most of their lives, right? Because otherwise, you can't have your immune system wreaking havoc on your primary nutrition source. So their immune systems have been weakened a little bit. So they also found more than expected transposable elements in their genomes. We've talked about this before. Those are snippets of ancient viral DNA that have become integrated into our DNA. I mean, basically all animals have that. We do as well. This is also mobile DNA and that's why they call it often jumping genes. You probably have heard the jumping genes name for these. This plethora of transposons, as they're called, makes sense as well since if you weaken your immune system, then you're going to be – you can expect more viral attacks that would then become integrated into your genome. So regarding that specifically, Lee said, these aspects highlight the tradeoffs of symbiosis. The host has to accommodate a suppressed immune system and potentially more viral genome invasions. Okay. Now, these – Steve, you're going to like this. These tradeoffs of photosymbiosis seem worth it. Check this out. The solar harvesting efficiency of the algae is actually pretty amazing. So for example, take farm crops. They capture approximately 3 percent of the incident solar radiation, right? Then you've got large solar arrays which can grab – what's the number now, Steve? 20 to 25 for the best commercial ones right now?

S: Yeah. The best commercial ones are around 24.

B: Okay.

S: 22 to 24.

B: So that if you consider it now future efficiencies with materials like pervascite, they're saying, oh, we can get up to 40 percent of – 40 percent efficiency. So I saw in another study about giant clams and they were saying that the algae on the giant clams – this is I think specifically the one that Steve and Jay and I saw. They're saying that those algae can get efficiencies as high as 67 percent, 67 percent. I checked out the study, Steve. It seemed totally legit. I didn't imagine it was quite that high. But they're trying to figure out – in that other study which I didn't study extensively, they were just trying to find out how are they doing that. They came up with some various models and ideas of how that's happening and potentially we could incorporate that into our designs as well. Imagine getting up to 60 percent, even 50 percent would be amazing. This isn't just – oh, look, this is interesting genomic research. The better the giant clams are understood, the better we can understand marine ecosystems themselves. Senior paper author Jinchun Li said giant clams are keystone species in many marine habitats. Understanding their genetics and ecology helps us better understand the coral reef ecosystem. And of course with climate change, unfortunately, and overfishing, giant clams have also been impacted as well. The T. maxima clams in this study, they're actually pretty good right now. Their current classification is of least concern. So they're good. There's so many of them. They're very widespread. So they're not really – they haven't been impacted too much. But my beloved Tridacna gigas, the biggest ones, the largest clams are now listed as critically endangered. That's about as high as the levels go before a species becomes extinct in the wild. So that's horrible news. Critically endangered. Crap. That kind of really sucks. So I hope some serious efforts are being made to prevent that. And of course all the other creatures that are also endangered because of climate change and overfishing and all these other things that are happening. So good luck to my buddy, the giant clam in the ocean by Australia.

S: I remember that too. I remember putting my hand in there and it closing rapidly on it.

B: Rapidly. And imagine you have no idea what was going to happen.

S: It snapped shut. Yeah.

B: Yeah.

S: But they are beautiful.

B: Oh my god, just magnificent. And gargantuan. That one probably –

S: Huge.

B: That was at least three or four feet. Right?

S: Oh yeah.

B: That was as big as – and that thing could have been like 500 pounds.

S: Yeah.

B: A quarter ton.

Who's That Noisy? + Announcements (1:13:31)

S: All right, Jay. It's Who's That Noisy time.

J: All right, guys. Last week I played this noisy. [plays Noisy] What do you think, guys?

C: Bird?

E: Was it two different things or was it one thing?

B: It sounded like a dolphin at the end there.

E: It sounded like two different animals.

C: Bird talking to a dolphin?

J: Not a bad guess. Well, we have a listener named Stavis Maples who wrote in and said, love you all. This is someone saying hello to a fox or coyote and they respond.

E: Oh, so we know what the fox said.

J: Yeah. So that's an interesting guess. Not correct but still fun. A listener named Scott Whitaker wrote in and said, hi, Jay. That clicky gurgle at the end of the call sounds like – a lot like a dolphin to me. So that's not a horrible guess, Scott. Not bad. A listener named Aaron Allison said, hey, Jay, I think it's a marine mammal, like a beluga whale. They're cute. And then he goes on to talk about how skepticism is needed because of the politics and all that. OK. But bottom line is that was a close guess. I did get 3 or 4 correct guesses. I'm going to list the first two. So the first person that wrote incorrectly was Abigail Weismer and Abigail says hi Jay and rogues all the way from Israel. I believe this week's noisy is a talking whale, specifically an orca called wiki. Thank you for your company on long drives for the past 18 years. And the second person who was right behind Abigail is Mike Nelson. He guessed correctly. So guys, this is a this is an orca that has been they found out that it was trying to basically mimic human speech and then it was successful. So listen to this. [plays Noisy] I mean, that's pretty good.

E: Yeah.

B: Yeah.

J: I like how it says hello, then it geeks out afterwards. It's like, hello. That's so awesome. You know, it's so funny. Instantly when you hear an animal mimic human speech, you can instantly feel your brain go, they understand. You know what I mean? It seems like he certainly knows what that word means, right? I would highly doubt it. But still, these are intelligent creatures, man. They have very complicated behaviors. And they're the apex animal. They have every right to be able to think clearly and speak. Anyway, that was really fun. That was one of my favorite noisies of the year so far. I have a new noisy for you guys sent in by a listener named PK. And here it is. [plays Noisy] All right, guys. So if you think you know what this week's noisy is or you heard something cool, always email me at WTN@theskepticsguide.org.

So I'm going to cruise through our announcements. If you've been listening recently, you've heard some of these before. First and foremost, I think the most incredible thing is that Steve is going to begin working for the SGU full-time probably early July. Right, Steve? You're retiring end of June.

S: End of June. Yeah, it'll be my last day at Yale. So yeah, as of July 1st, I'll be working full-time for the SGU.

J: So if you want to help us to support Steve's move to the SGU, and you also want to support the absolutely important need for critical thinking and logic and skepticism in today's world, please consider becoming an SGU patron today. You could become a patron at any level. There is no requirement. Any amount would be helpful for us. So it is really a great time to join the SGU. It's also a great time to become a patron of the SGU and then come to NOTACON because NOTACON is about listeners of the SGU getting together, socializing, making friends, having a wonderful weekend. NOTACON will be on May 15th, 16th, and 17th. You could go to notaconcon.com for all the details. I have been promising to put the schedule up. Well, we have finalized our list of events and I am putting them in order right now. So probably next week it'll be up for your perusal. We hope that you join us. A couple of interesting call-outs. This year we're going to be doing a Beatles sing-along that George Hrab is going to host. This will happen on Saturday night. That was fantastic. The last time we did 80 sing-along and people dressed in costumes and everything. It was just a blast. We have some new bits that we're going to do this year. Was there anyone in particular that you like best, Steve?

S: I mean, I think I'm most excited to do the SGUniversity again because it's new topics.

J: Which is what?

S: SGUniversity. We each give a 15-minute talk with as much audience participation as possible, like a workshop on something that we know that's outside of the realm of skepticism.

J: Like a skill set or something.

S: Something. Like last year, Evan did how to make a board game. And that was really popular. I did how to do a neurological exam. This year I have something I'm very excited to talk about planned.

J: So if you're interested, guys, to join us at the conference, go to notaconcon.com for all the details. You could join our mailing list. We have consistently been sending out an emailer every week outlining everything that the SGU has done the previous week. It's very easy to sign up. Just go to the theskepticsguide.org homepage to find the link there. And you could give our show a rating, if you don't mind, to give us a few minutes of your time. This will help new people find the podcast. And that's it, man. It's a wrap. I have one more thing to say, Steve. I am planning all of the live shows now for next year. I got our first proposal in. I expect a few more to come within the next week or two. I'll be sharing all these potential dates with you guys. And I'm just super excited. We're going to be moving around the country.

S: All right. Thank you, Jay.

Emails (1:19:36)

S: We had a couple of emails on the telepathy tape. So I just wanted to do a follow-up there. So last week, Evan, you talked about the telepathy tapes, essentially this podcast talking about doing facilitated communication with children who have an impaired ability to communicate. And some of the practitioners have come to believe that the children that they are facilitating not only can communicate far beyond any objective neurological assessment would tell you, but that they are also telepathic.

E: That's a new level, right.

S: They could read minds. And so some of our listeners thought that we didn't go into enough technical detail on that. So this happens from time to time. Because we have such a back catalog of shows, 20 years of shows, oftentimes we will self-reference. So we talked about the fact that, well, this is facilitated communication. And we've spoken about FC many times on the show before, given a description of what it is exactly and the fact that it's complete and utter bunk, that it has been disproven. And so we sort of rely on that, that people know that, or they could certainly look back into our back catalog, or you could always look on one of my blogs, either Science-Based Medicine or Neurologica, and I guarantee you anything we talk about like that on the show, there will be one or more, often many articles doing significant technical deep dives on those topics there. But because people ask, we have to also remember to balance that with the fact that we have lots of new listeners, they may not have, not everybody has listened to all 1000 episodes of the show. And it's always hard to know, like, how much do we need to get to recap stuff we've talked about multiple times in the past. But I just thought, since somebody brought it up, say, okay, let me just give you the primer on facilitated communication, just very, very quickly. In the late 80s, this technique came out, essentially, what how it works is that the facilitator will hold the hand of the client. And while they're being asked questions, or whatever they're being communicated with, and the client will communicate by pointing at letters on a letterboard, for example, or they might be hitting keys on a typewriter. And when this came out, many people who work with children who are non communicative, thought it was a revolution, like, oh, my goodness, there's much more of an intellectual life going on inside these kids brains than we thought. There was more of a problem with communication than a cognitive problem. And now we have found a way to break through and to communicate to these kids. The whole thing imploded within a few years, because when you actually subject the technique to objective testing, you subjected to double blind testing, it turns out that the children were not doing the communicating that the facilitator was doing 100% of the communication. And there were many videos were shown showing how implausible the claims were, oftentimes, the children were not even looking at the board or the keyboard, which is impossible. You know, you can't blind type, even somebody who is neurologically typical and intact, can't do that, right? With one finger, you could do it if you have your keys placed on the home keys where you know where they are, but you cannot one finger type where you have no reference. It's just not possible to do that. I've also seen videos in this, some newer ones that we've talked about on the show more recent ones, where the person being facilitated is typing really fast, like they're doing a pretty decent job of typing pretty quickly. Now, of course, the facilitator is looking intently at the keyboard, the client may or may not be looking in the direction of the keyboard. So imagine what they're claiming, that this person is directing the facilitator to the key they want to hit. And they're doing it rapidly and precisely, again, impossible, impossible for even a neurologically intact person to do. So we are being asked to imagine that children who have profound neurological impairment have multiple skills that go way beyond even an average or a typical person. And then it keeps getting worse because again, once you disconnect the communication from reality, like the child is not doing the communication, the facilitator is, that's been proven. Then you can make it seem as if that child has all kinds of abilities. So I've, again, been directly involved in cases where it was claimed that a 10-year-old was reading on a 16-year-old level. And even despite the fact that they've never been explicitly taught how to read, they're not only reading, they're reading at an advanced level, as if there's some kind of savant, super genius in there. And also they speak other languages that maybe they've only been peripherally exposed to but not explicitly taught. Like again, showing superhuman cognitive ability, not just, oh, they can sort of communicate if you sort of eliminate this physical limitation. It's like, oh my God, they're superhuman. And then, of course, they're also telepathic is now the next layer, because of course they are. They're whatever it is you test them for, because the test itself is broken. And this is a good skeptical lesson, is like when everything seems to turn out positive, it's not that you've hit upon some miracle, it's that your assay is probably broken, right? It was like the homeopath who thought he found the cause of all disease, because every slide he looked at had these little oscillococcinum on them. Well, they were air bubbles contaminating every slide, right? It was a contaminant, but he thought he found the cause of disease. And then of course, it was on every slide. So we found the cause of all diseases, or it's an artifact, right? So again, if the child is demonstrating multiple, multiple superhuman abilities, it's probably probable that your technique is artifactual. It's not working. People who were responsible, ethical, mature, in my opinion, and scientific, were able to accept that it was self-deception and move on with their lives. But some people have not been able to do that. They are too invested in the notion that these children are talking, that they have this rich inner intellectual life, and that this fantasy really that they created is the person. And so they essentially steal their identity. They steal their voice. They subject them to things that are not in their best interest in order to feed this FC-fueled fantasy. I've seen instances where they went to college, like with their mother facilitating them. Just ridiculous.

C: There's a crazy documentary where the facilitator falls in love with the person.

S: Oh my goodness. They start to project onto it. There are also people sent to jail based upon accusations that came out under facilitated communication. That's the really dark side of this, is when the wife is facilitating the child who then accuses the estranged husband of abuse. And the husband then goes to jail based upon that testimony. Absolutely unconscionable. And it's the profession that fell for this. This is completely on them. They did not do their skeptical scientific due diligence before rolling this out. And then, okay, fine, within a few years, it sort of collapsed. They should have just said, okay, that was our bad. We should have been more skeptical upfront. But now we've learned our lesson. And many did do that. But for those who didn't, who doubled down and tripled down on this pseudoscience, there is no excuse. This is unprofessional. It's unscientific. It's unethical, in my opinion. It's a complete failure. And they are continuing to inflict this on children, on families, on anyone that it touches. And it is really a scandal. Right? And it's been, that's what we're dealing with here with the telepathy, with the telepathy tapes. They're adding this new element of, oh, yeah, and they're psychic, too. Yeah.

E: Yeah. And they're appealing to a younger generation of listeners who don't know. And this podcast is perhaps their first experience being introduced to this subject at all. They have no idea the 30, 40 years of background into this.

S: Yeah. All right. Let's go on with science or fiction.

Science or Fiction (1:28:45)

Theme: The Moon Item #1: The Moon is the densest moon in our solar system. Item #2: A recent analysis finds evidence of geological tectonic activity on The Moon as recently at 160 million years ago, suggesting it might still be active. Item #3: The Moon has a weak magnetic field, measured at the Apollo 16 site at 0.31 microtesla (compared to Earth’s 50 microtesla field).

Answer	Item
Fiction	The Moon is the densest moon in our solar system.
Science	A recent analysis finds evidence of geological tectonic activity on The Moon as recently at 160 million years ago, suggesting it might still be active.
Science	The Moon has a weak magnetic field, measured at the Apollo 16 site at 0.31 microtesla (compared to Earth’s 50 microtesla field).

Host	Result
Steve	win

Rogue	Guess
Cara	The Moon is the densest moon in our solar system.
Bob	A recent analysis finds evidence of geological tectonic activity on The Moon as recently at 160 million years ago, suggesting it might still be active.
Jay	The Moon is the densest moon in our solar system.
Evan	The Moon is the densest moon in our solar system.

Voice-over: It's time for Science or Fiction.

S: Each week I come up with three science news items or facts, two genuine and one fictitious. And then I challenge my panel and skeptics to tell me which one is the fakearooney. There is a theme this week. Well, these are all kind of based on recent-ish news items, but there's a theme. The theme is the moon. How much do you guys think you know about the moon?

E: Which moon?

C: Not a lot.

S: The moon.

C: Our moon?

S: The moon. I like taking topics that you think you know a lot about and going a level deeper, because no matter how much you know about something, there's always more details. All right. Here we go. Item number one, the moon is the densest moon in our solar system. Item number two, a recent analysis finds evidence of geological tectonic activity on the moon as recently as 160 million years ago, suggesting it might still be active. And item number three, the moon has a weak magnetic field measured at the Apollo 16 site as 0.31 microtesla compared to Earth's 50 microtesla field. Cara, since you profess to know not much about the moon, and it's a welcome back, you get to go first.

C: Oh, boy. Okay. The moon is the densest moon in our solar system. Well, there's lots of moons in our solar system. What are the odds that our moon is the densest? I don't know. No idea, like, what makes a moon denser. Well, I mean, its density makes it denser, but I don't... Anyway. A recent analysis finds evidence of geologic tectonic activity on the moon as recently as 160 million years ago, suggesting it might still be active. Eh, isn't it full of craters? Those aren't all impact craters, right?

S: So I'll just clarify for you, yeah, tectonic means not from, like, impact.

C: Oh, you're saying specifically, like, plates.

S: Tectonic activity, internal geological activity.

C: Right.

S: That's why I threw that word in there.

C: Okay.

S: Yeah, it's getting hit by meteors all the time. That doesn't count.

C: Does it have tectonic plates? Well, I mean, whatever. It's got activity. Does it have rifts and bowels? I feel like when you look at the moon, it's rifty. Yeah. I don't know. That could be true. Maybe not.

B: Steve, do you want to, like, change the word tectonic there?

S: No, I don't want to change the word tectonic.

E: Ooh.

C: Yeah, because tectonic refers to the movement of plates, right?

B: Yeah.

C: By definition.

S: Maybe.

E: We will find out soon.

C: Uh, okay. Well, you did just say, like, under, like, geological activity. Like, beneath, or, like, deep geologic activity. But you're not talking about-

S: It's internal. It's not external. It's not from impacts. It's from internal geological activity.

C: Right. So, it could be volcanic, and it could be plates moving around, and it could be all sorts of things.

E: Yeah, it could be moon creatures digging holes.

C: Sure. Could be.

E: Could be. We don't know.

C: There's those lava tunnels, right? Hey. That's gotta be something.

B: Yeah, baby.

C: So, yeah. So, maybe there is some chance that there's still some activity. I don't know. Geology's weird like that, man. There's, like, these dormant volcanoes, and then they're like, whoa, it came alive again. And then it's got a weak magnetic field measured at the Apollo 16 site at .31 microtesla compared to Earth's 50 microtesla. Sure. Yeah, I feel like from a statistical perspective, I'm gonna say that it's not the densest, because there's, like, hundreds, maybe thousands of moons. Are there moons that we haven't even identified, like, way out in the Oort cloud? I don't know. So, at least hundreds of moons in our solar system. So, I don't know. What are the odds that it's the densest of all of them? Probably low. So, I'm gonna say that that's the fiction.

S: Okay, Bob.

B: Yeah, the density, I just don't know about the density of the moon. I know it's probably more dense than icy moons, but other than that, I just don't know. Let's see, the second one. This geological tectonic activity, I don't know if Steve's just trying to save his butt there and say, no, I meant to say tectonic. For me, a tectonic plate is an external plate floating on a more liquid interior, like we have on the Earth. And so, I don't know what to think about that one. Could there be some internal geological activity? I've heard hints of that over the years. It's looking solid, though. And then this weak magnetic field, I mean, I'm really straining my memory here to remember if there was a weak field. There might be, but wouldn't that field imply some internal geologic activity, like by definition? You would have to have something internal going on in order to generate that magnetic field, unless there's just some other subtle process about with radioactivity. All right, so, because I just don't know, I'm not confident about any of this in terms of what the answer definitely is. But since two and three are somewhat related, like in my mind, if two is true, then probably three would be true, and vice versa. So because of that lame connection I happen to make here, I'll go with Cara and say that it's not necessarily the densest moon. There's so many of them. You might see some having just a denser rock, and you wouldn't need much of it. I mean, these are small moons out there that wouldn't need a lot of that dense rock to have greater density than the moon, so I'll go with one fiction, density of the moon.

S: Okay, Jay.

J: Yeah, I'm saying I'm agreeing with Bob and Cara. I don't think our moon is the densest. When you say density, Steve, you're talking about its size versus weight. I'm just thinking out loud here.

C: Volume, yeah.

B: Mass per unit volume.

C: Mass over volume.

B: You're my density.

C: Drop it in the bathtub.

S: You're my density.

J: So the moon...

E: Back to the future.

J: I don't think that the moon... I'm pretty sure the moon has a magnetic field, and I think it's super weak because it doesn't have what Earth has, which is we have an iron core to our planet. That's what generates our magnetic field, and that it's moving, right? I know that the moon is important. I don't think the moon really has any of that.

C: Dynamo.

B: Exactamundo, Cara.

C: Yeah, I learned something on SGU.

J: Yeah, I think that's basically it. I mean, I agree with these guys. I think the moon is not the densest.

S: All right, and Evan.

E: If the moon is the densest moon in our solar system, that is one dumb moon. Thank you. So for all the reasons stated by my co-hosts, I am in agreement. And Bob, yeah, I also connected two and three together. So yes, I was thinking the same. And also, what are there, 100 moons that we know of in our solar system? So numerically speaking, the density item here should statistically be the most likely fiction.

S: Okay. So you guys all agree on number one. Let's start with number three. You seem to have the easiest time with that one. The moon has a weak magnetic field measured at the Apollo 16 site at 0.31 microtesla compared to Earth's 50 microtesla field. You guys all think this one is science. And this one is science. Although just about everything else you said about it is wrong, but let me go over it really quick.

B: Who? Me?

C: As is our way.

S: It is still controversial whether or not the moon ever had a magnetic field generated by an internal dynamo. Because it does have an iron core, but that iron core might have been too small to generate a magnetic field. But it's still possible that it did generate one early on, like 4 billion years ago, early on in its life.

E: Proto-moon, right?

S: No, no. When it was still a moon. When it was a moon, it was a moon-moon. But before it cooled to the point where it no longer would have had an internal dynamo.

E: And it was so big in the sky back then.

B: So big. 15 times bigger than it appears now. Now-

C: Oh, that's cool.

S: But so the moon's current magnetic field definitely is not created by a dynamo. No question. It's not even on the table.

E: So what else creates magnetic fields?

S: What else creates magnetic fields, right? So it's basically created in the crust itself, right? So there's iron in the crust is creating the magnetic field. So how did the minerals in the crust become magnetized? So I said measured at the Apollo 16 site. If you measure it in other locations, it's different. So it's a variable magnetic field.

B: Well, Mars is like that.

S: Similar. But this is all, the moon has a magnetic field pretty much all over. It's just highly variable. So for example, at the low end, it measures 6 nanotesla.

B: Nano. Tiny.

S: That's tiny tiny. So, but there's two theories as to how the minerals in the crust got magnetized to the point where they're still creating a measurable magnetic field. One is that those magnetic fields were laid in billions of years ago when the moon did have a dynamo, right? So it had a strong magnetic field. It could have been super strong, like even stronger than the Earth today, like two or three times stronger than the Earth's current magnetic field. And that that induced magnetism in the crust, which survives to this day. But there are lines of evidence against that. Like there are things we should have seen on the moon that we did not see in terms of, like it should have induced this change in this mineral and it didn't. And so it probably didn't have a dynamo magnetic field, but it's not clear. It's controversial. So the alternative theory is that when meteor, meteors impact the moon.

B: They were magnetic already.

S: No, but that induces magnetism by the impact. The energy of the impact induces the magnetism. And there is some evidence to support that hypothesis.

B: That's weird.

S: So yeah, so that's the two hypotheses as to why we have sort of a crustal magnetic field on the moon today.

B: Interesting.

S: Yeah. All right. Let's talk about number one a bit here.

J: Number one.

S: Number one. So you guys think it's just statistically unlikely that the moon is the densest moon out of all the moons in the solar system. There are 293 moons.

E: Oh, it's even...

S: As of January-

E: ...more statistically...

C: That we know about.

S: -that we know about. Yeah. So NASA estimates there are 293 moons orbiting planets in our solar system. But they said there are likely more moons to be discovered. Now there are a couple of things that are unique about the moon. One is it's the closest moon to the sun, right?

E: Correct. Venus has zero. Mercury has zero.

B: So less ice, more density.

S: So less volatiles, right? So anything in the outer solar system with volatiles is going to be less dense, pretty much. Right?

E: You would think so.

B: Yeah.

E: Unless it was captured or something.

S: Number two, right? Our moon might have a unique origin.

B: Yeah. Yeah.

S: The other planet hitting the proto-Earth, throwing the debris up. Did you know that the inner planets are all denser than all the moons?

B: Oh, that doesn't bode well then because, yeah, there's a lot of Earth mark one in the moon.

S: Exactly.

E: Called the rocky world.

S: The moon-

B: Make it denser.

S: -is more characteristic of an inner planet than any other moon. And those inner planets are denser than all of the moons.

E: It has to be with the in-crowd, you know?

S: So it's actually not statistically remarkable that the moon would be the densest moon in our solar system.

E: I was only figuring it by number, pure numbers.

S: Right.

C: Me too. Yeah.

S: And, in fact, for a time, it was the densest moon in our solar system.

C: But not now?

S: Until...

E: Until...

S: They discovered that Io is denser.

E: Io.

S: So it's the second densest moon in the solar system.

E: Awesome.

S: Io, as you may or may not know, also has no volatiles because this is the volcano planet moon, right?

E: Yes.

S: It's constantly turning itself inside out with geological activity because of the tidal forces from Jupiter. So it's also extremely dense, just a little bit denser than our moon. So, yeah, so the moon is the second densest object in the solar system.

J: Nice.

S: Right. But not statistically unlikely because of those... for those reasons. Yeah, so if Io didn't have the unique configuration that it does, it would have been the densest moon in the solar system.

C: And we would have gotten it wrong.

S: That's right. All right. Let's go on. Number two, a recent analysis finds evidence of geological tectonic activity on the moon as recently as 160 million years ago, suggesting it might still be active, is science. So Bob, I wouldn't throw in the word tectonic there without making sure that that was correct.

B: I hope so.

S: It's a language used by the researchers themselves. I think you are thinking that tectonic refers only to tectonic plates. I think the term is more generic to any kind...

C: I just Googled it and every definition says the process by which the Earth's crust... Like tectonic was named for plate tectonics. So maybe the word's been co-opted since then.

B: I think if it's not paired with the word plate, then it kind of...

C: No, no. If it does look...

S: Because there are lunar tectonics, there are contractional tectonics. So tectonics is just dynamic change in the crust, right? And so, or in the layers of a world, of a planet or a moon, it doesn't have to be plates moving. You can have other types of tectonic activity. So we know in the past there was tectonic activity on the moon. The thinking was that it stopped a couple of billion years ago, two and a half billion years ago when the moon cooled, right? The moon cooled to the point where the crust basically solidified and the lower layers also were too cool for there to be any significant activity. There were volcanoes on the moon way in the past, nothing for billions of years. But a recent study found that there are... They found on the dark side of the moon, the darkly colored side of the moon, or the far...

B: Good save.

S: The far side of the moon, the side of the moon where you have the dark maria, right? I know it's kind of like the... It's funny because people think that there is no dark side of the moon. It's like, you're right, but there is a darkly colored side of the moon. So it's kind of...

B: How about the lower albedo side of the moon?

S: So on the far side of the moon, they found these structures, these ridges, it's called small mare ridges, which are not caused by impacts. They are similar to ridges that are seen near ancient volcanic activity. So the analysis shows they are probably tectonic in origin and they crater age them. So you age them by counting how many craters there are. Because since there's no atmosphere, there's no erosion, how many craters there are on the surface material on the moon is a pretty good estimate of how old it is, right? So the older it is, the more craters there are. And they age this as being way younger than they would have thought, like there was any kind of tectonic activity on the moon, it was just 160 million years. And if there could be tectonic activity that recently, it could still be going on today.

E: Oh boy, we're going to maybe find out more about that very soon.

S: Right. I mean, we know there are moon quakes. There's some geological activity going on on the moon. So again, this is one study. It found this, obviously everything is subject to revision, but that was an interesting finding. Yeah. And I was a little bit surprised at the use of the word tectonics myself when I was reading the study, but I had to read enough to say, okay, they're meaning it as a more generic term.

B: Yeah, this will really throw Bob for a loop.

S: Well, I also, I put it in there so that you wouldn't think it was just meteor impacts, right?

B: Right.

S: Even though I still had to clarify that. That's why I put that in there. It wasn't to confuse you too much as to not make you think, oh, sure, stuff's hitting the moon all the time. It's like, no, no, no, that's not what we're talking about here. We're talking about tectonic activity.

E: Should you have used the word geological though?

S: They used tectonic in the study.

E: Yeah, but the word geological, isn't that specific to Earth only?

S: No, I don't think so.

B: No, I don't think so.

S: Nope.

E: I think it is. But geo means Earth.

S: No.

B: Terrestrial.

S: No, and they used the word geologic in the study too, Evan. I'm just using the terminology they're using.

E: I understand, but they could be wrong.

C: But I'm with Evan.

E: They could be wrong.

C: I'm with Evan. They're co-opting all those terms.

E: Yeah, they should be-

C: All those terms started in reference to Earth, and their first definitions are all having to do with Earth.

S: Right. Most planetary geological stuff are analogies to stuff happening on the Earth. Pretty much everything was first named as an Earth phenomenon, like volcanoes or whatever, quakes. Everything. There aren't earthquakes on the moon. There are moonquakes on the moon, right? Which always reminds me of-

B: Yes. Flash Gordon.

S: Flash Gordon.

E: Flash Gordon, right?

S: At the beginning of the movie, he's got this board where he could press buttons and make environmental catastrophes happen on the Earth.

B: This is an alien business.

S: This is an alien. He's on an alien world. One of those buttons is named Earthquake. Not moments after we see the button labeled Earthquake, his lackey is saying, yes, there's a small planet here that refuses to pledge whatever they're fealty to you. The locals call it Earth. Earth, you mean like that button on your board called Earthquake? That Earth?

B: What a coincidence.

E: It's the only problem with that movie.

S: It just struck out to me for some reason. I've never forgotten.

B: We should watch that. I mean, it's literally been three dog ages since I've seen that.

E: That would be fun. We could do a burial of that movie easily. My friends and I buried Star Wars Episode VIII recently. We watched it together. We took a pledge that it will be the last time we watch it, and we just roasted it the whole time. It was wonderful. So cathartic.

S: Episode... the VIII movie?

E: Yeah, VIII movie.

S: Oh, God. It was so bad.

E: Oh, yeah. We buried it.

S: You know, I was holding out a little hope after VII. It's like, OK, a little bit of a rocky start here, but they can pull this out. I could kind of see where they're going. There's lots of interesting ways they could go here, and it just went downhill. If you know anything about the history of how that was written, it was a CF. It was a complete CF. Nobody was in control. Anyway, I don't want to get into this. It was just an abomination. Evan, why did you bring this up? All right.

E: You're welcome.

Skeptical Quote of the Week (1:48:32)

“Our beliefs do not sit passively in our brains waiting to be confirmed or contradicted by incoming information. Instead, they play a key role in shaping how we see the world.”  – ― Richard Wiseman, (description of author)

S: Evan, give us a quote.

E: "Our beliefs do not sit passively in our brains, waiting to be confirmed or contradicted by incoming information. Instead, they play a key role in shaping how we see the world." Richard Wiseman, who is a wise man.

B: Richard!

E: He's awesome.

S: Nice quote. Yeah, that is a key, I think, psychological finding, the idea that, yeah, we are not just passively collecting information, perceiving information, remembering, processing, thinking about it, whatever. It's an active narrative process, right? The narrative dictates the facts more than the facts dictate the narrative, unless you take a scientific approach. The whole point of science is to reverse that causation so that facts dictate narratives. Otherwise, instinctively, just psychologically, we impose our narratives on the world, not the other way around. All right. It's a good quote.

E: Thanks.

S: Thank you all for joining me this week.

B: Sure, man.

E: Thanks, Steve.

C: Thanks, Steve

Signoff

S: —and until next week, this is your Skeptics' Guide to the Universe.

S: 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@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.

[top]