Hosted by Lester Nare and Krishna Choudhary, this rundown episode covers five new science and tech stories at a high level: NASA’s Artemis 2 moon mission, what actually leaked in the Claude Code incident, a new cancer genomics paper suggesting domesticated cats may be unusually useful real-world models for human cancer, two leaked iPhone spyware toolkits, and a science-focused review of Project Hail Mary. Summary Artemis 2 is finally flying - why this mission matters, why it is not landing yet, and why the moon race is back in geopolitical focus. Claude Code leaked, but not Claude itself - what was exposed, why people got confused, and why the distinction between source code and model weights matters. Cats and cancer - why domesticated cats may offer a more realistic environmental cancer model than traditional lab rodents. iPhone spyware in the wild - what Dark Sword and Coruna are, what they can do, and why this signals a broader shift in cyber risk. Project Hail Mary science review - what the film gets right, what it gets wrong, and which scientific liberties are hardest to buy. Support the show Donate: FFPod.com/donate Follow: @FFPod on X / Instagram / TikTok / Facebook
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
Transcript
From the show. Tap a time to play from there.
So Microsoft products, they're the bane of everyone's existence at work. Right. And that work also includes astronauts going to the moon. Okay, so did you see this leak story about Claude Code? Yes. So do I just have Claude's LLM now or what? Not quite, but let's understand what actually happened. So your cat could be the key for cancer research in humans. Very interesting. Project Hail Mary. Everyone's complaining about the centrifuge scene. I think there's something even deeper that's wrong. Okay. Hello, Internet. This is your captain speaking, Lester Nare, joined as always by my co-host and our resident Ph.D., Krishna Chowdhury.
Welcome back to another rundown episode where we're going to just cover five new both science and tech stories this week, just at a super high level. And we have some some interesting stories to cover. We're going to start off with the Artemis launch that took off yesterday after several delays. It was on April 1st, but this is not happening. An April Fool's joke. It definitely did take off. We're going to follow that up with the very viral leak story about Claude, the anthropic model leaking. Oh, my God, does this mean we get to use AI for free?
We're going to talk about what exactly happened, why. There's some interesting subtle details about what actually transpired with the Claude leak that happened earlier this week. There's a major new science paper that suggests that cats – Maybe one of the best real world models for sort of studying cancer in humans for an interesting reason about location. There's a trend you may be sensing because we're going to have another leak story about an iPhone spyware tool, two of them actually, that have leaked onto the public internet.
And this is sort of this trend of location. more nation state level elite spyware tools now becoming more accessible commercially. What does it mean for you if you're an iPhone user? And we will then wrap up with a quick review of the science and, in Project Hail Mary, the ultra-popular new theatrical release featuring Ryan Gosling. We will have two perspectives on the movie, which we both saw recently.
We're going to cover some topical points, but closely related to the science and first principles, because this is From First Principles. So yesterday, NASA launched Artemis II, the first crewed mission beyond low Earth orbit.
We've got our trusty Lego-built Artemis II SLS space launch system here that we built out of Legos. By we, I mean Lester did all the work, and I am going to take credit for it because that is our dynamic, right, as friends. Anyways, launch date, April 1st, 2026 on April Fool's. The launch window opened up. The Earth and the moon were in the right spot. So, you know, you got to do it. It's the first crewed mission of the Artemis system. They tested the SLS rocket and the Orion spacecraft.
Right now, as we are taping this, this is on Thursday, April 2nd, they have already completed their apogee rays burn, and they've done that high elliptical orbit where they go a little bit far away from the Earth and they test all those systems. Now they're on their way back, and if everything goes well, then tonight, in about three hours, they are going to go to the moon. They're going to... put their rocket launchers on, and go all the way to the moon. I am really excited. The crew, four veterans of space.
Actually, no, I should say three veterans of space travel and one rookie. The three veterans are Reed Wiseman, Victor Glover, and Christina Koch from NASA. And the rookie is Jeremy Hansen from the Canadian Space Agency. Can you imagine? It is his first... It is his first time in space, and he is going straight to the moon. It's unbelievable. What a lucky guy. Yes, unbelievable. Yeah. And I think we have some good graphics here for what it's going to look like. Yes. I wanted to actually go through some of the graphics of how the Artemis mission is going to go to the moon because I think it's like –
a really nice exercise in orbital dynamics physics. So if we go to the first graphic, this is from NASA. They're showing that Artemis does that apogee rays burn, and then it slingshots around Earth and goes to the moon. Now you can see the moon is still traveling with the Earth. It takes three days to get there. The Artemis is going to meet the moon, slingshot around Earth, and then return back to Earth. What I like about this is this is an Earth-centric view. Okay, so the moon is orbiting around the Earth, and Artemis is going to the moon and back, right?
And the camera is sort of moving with the Earth. Now, the Earth... Yeah. What looks like it to me, when you first look at this thing, right, Artemis is going there, and then it just turns around. Yes. And if we notice, it seems like the moon is doing so much work to turn this spacecraft around and head it back to Earth.
Right? Now let's look at our second video, and that is from the moon's position. So now the camera is moving with the moon, and you can see now Artemis is approaching. It's approaching. It's approaching. And it gets slightly deflected. Mm-hmm. The moon is actually not doing that much work. So how do you reconcile these two camera angles? From the camera angle of the Earth, it's like the moon just slingshotted it back. But from the moon's perspective, there's just a tiny bit correction. That kind of makes sense because the moon is not that big.
So what exactly is happening? We've got to understand that when Artemis is going towards the moon, it does a single burn... And then it launches towards the moon, right? When it does that, it does that single burn, and then it turns off its engines. So it's still falling towards the Earth. It's like we threw a baseball up, and that baseball is going to have a really high velocity, but it's going to keep falling and reducing its velocity because the Earth keeps pulling on it.
So if the moon wasn't there... Artemis would go all the way to the moon, go way over the moon's limit, but still return back to Earth. Because Artemis does not have escape velocity from the Earth. There's a gravitational pull. Exactly. So when we go back to the moon's perspective... Now you see why the moon, even a tiny bit of deflection, is enough to do that slingshot maneuver, right? Artemis is falling towards the Earth this whole time. All the moon is doing is just nudging it a little bit more to fall towards the Earth as Artemis goes around.
It's kind of an interesting way to think about perspective in space, right? It's like Artemis would still return back to Earth, but the moon is just there to kind of have that slingshot. Just a tiny bit more. So one of the things we covered in our Artemis episode that we previously did a deep dive on. So if you want more information on this, we've actually covered this previously. But the launch is super relevant. And I know we have one more graphic to show here. So the first major question you see in every comment thread on every NASA account is, oh, we're going back to the moon, but why aren't we landing?
And this is something we did cover at length in the long form. We did Apollo in the 60s. We have not done anything similar. The first point I want to bring up is that low Earth orbit and going to the moon is the distances are just fundamentally very different. Oh, yeah. Insane. We've got 200,000 miles to the moon and low Earth orbit is like 100 miles. Right. Tops. Right. Right. So it's just these are just different.
Yeah. Like several orders of magnitude. Yeah. Right. You can run. If you can be good at a 5K, that doesn't mean you're going to be good at an ultra marathon. Yeah. You need different training. Yeah. You need different diet, body mechanics, all these things. And this isn't even a 5K. I think it's closer. Like, it's like a 400 meter relay versus a marathon. Right. Right. Right. Right. So pretty insane. And part of the reason why, again, we're not landing is the analogy I brought up when we did the long form episode is,
If you're the Dallas Cowboys, the football team, and you won so many Super Bowls decades ago, that doesn't mean you're going to win Super Bowls today. No. You have different personnel, different coaching, different financial incentives, and knowledge is lost when you don't continue. You have dynasties because you have the same coach coming back with the same training staff, the same offensive, defensive coordinators, and you're able to build that muscle memory. When you have decades that go past, we just don't actually have... The tools. Also with Apollo at the time, we did several of these exact same types of missions.
Yeah. Making sure we can get into lower orbit, then making sure we can get around into the orbit of the moon. But always coming back before we ultimately landed the rover. And this time around, we're just doing literally... The same process. Yeah. It's why it's Artemis 2, because 3 and 4 are ultimately going to get us to that landing spot. While we relearn with newer tools, newer systems, a new gaffer coach, a new personnel, all these things. And just to continue on that point, with Apollo, right— Apollo 8 and Apollo 10 went to the moon and never landed.
Right. Because you got to make sure you can get there and also scope out sort of a parking spot. That's what they're doing here. They're going towards the back of the moon because they want to scope out a parking spot about where to land for Apollo 3. Sorry, Artemis 3. Maybe also check out what the Chinese are doing on the dark side of the moon, which we can't see. Right. From Earth. Right. Yeah. And just to linger on that point, actually, with the dark side of the moon, right, one of the big things is that Artemis is going to go around the moon and see that dark side, which hasn't been seen by astronauts in a very long time.
Right now, as of the taping of this episode, so Thursday, April 2nd, one day after the launch, the moon has a full moon phase. The moon is a full moon in the sky, right? But what that means is the dark side of the moon is completely dark. Okay. In three days, it's going to be a waning gibbous, which means that most like, you know, 70% of the moon is going to be light for us. But when they get to the other side of the moon, it's going to be like pretty dark. There's going to be a line, though. Where they're going to see a crescent.
The moon is going to be kind of a crescent. Because part of it is going to be lit up by the sun. But the rest is going to be dark. And I had a naive question. Shouldn't you time it so that it's a new moon? That way the dark side of the moon is lit up. Right? I looked it up. It's a pretty clever... timing that they've done. The reason is, if you've ever been to the Grand Canyon, have you been to the Grand Canyon? I have not. Okay. But you definitely should. It's one of the greatest places on Earth, by the way. But if you go to the Grand Canyon, the best pictures are taken at sunrise and sunset.
Hmm. Okay. Why? Because there's shadows and shadows give you a profile of the canyon. That's actually okay. Okay. If you go when it's fully light or if you take photos of the Grand Canyon, when the sun is overhead, the photos don't look so good because you don't get that contrast between shadows and you can't really tell topographically the terrain that well. So they want to go when they're at a cusp between light and dark, because that's when we can see terrain really, really well. Especially if we're looking for a parking spot.
being able to understand what the surface looks like matters and this is a more ideal circumstance so in three days when they get there there's actually going to be a line they're going to see a crescent on the dark side and that's when they can really see craters and things like that really really well so I thought that was really cool the final thing that I wanted to talk about was the problems that they've been having I think their toilet broke but I think they fixed it so that's good because that would be definitely I want to go home right I like the moon.
The moon is great, but like, yo, the toilet is broken. Come on. No. I want to go home. That's tough. That's tough. I want to go home. When that happens at an Airbnb, I get a hotel. Yeah. Right? I'm not staying. And then I will petition Airbnb for a discount or whatever later. So they fixed the toilet. So that's good. The next thing that broke, though, was their Microsoft applications. So astronauts just like us use Microsoft and Outlook.
Yeah. Apparently was not working. If you go to the news, Artemis 2 astronauts find two Outlook instances running on their computers. And they had to call Houston to fix it because they couldn't open Outlook. And so Houston had to remote into their computer to figure out what was going on. I don't know if, like, NASA uses Teams. But anyone who's used Microsoft Teams knows that it is probably the worst thing. tool ever, but yet we're just stuck with it. And I guess NASA is also stuck with outdated Microsoft applications on a computer that's going to the moon.
Yes. I don't know that I don't think this is proprietary information, but I work at Disney Streaming, and they had to move from Slack to Teams. To Teams? Oh my gosh. You can imagine how fun that was. Oh, dude. Oh my god. It's incredible. Microsoft continuing to cause problems, even in space. Even in space. Even in space. We're trying to go to the moon, and Outlook? Outlook? Really? Another just brief fun fact that I'll touch on is a lot of people have been commenting on the difference in the quality of the media capture on the launch, right, versus SpaceX, which is in Blue Origin, which have been the commercial space that we've seen so far.
And this kind of just goes back to a couple of different things, right? The purposes are different. Funding is different. The sales aspect for the commercial airliners, space companies, matters a lot. Yeah. And so that is their product, is showing off the thing actually in space. They invest heavily in that. They also have now a network of stuff they can use up there to do so. And – When we're crunched for budgets in NASA, which have gotten even more crunched in the current administration, the money for the media budget is the first thing to go.
Yeah. And while you can focus on life support systems and things that are mission critical, safety, et cetera, et cetera. So this is why funding in science. You want good stuff? Tell them to stop defunding NASA. They also gave the... astronauts iphone 17 pro maxes okay to take photos while they're like like non science related just fun photos while they're up there as well so i'd be curious to see uh what those look like yeah i wonder if they have any photos of them fixing the toilet
That's got to be fun. I wonder who had to do it. I bet it was the rookie Canadian. 100%. It has to be. It has to be. You haven't been here yet. Yeah. This is your first time. This is your first time. You got to get a little hazing. Yeah. So it's great to see the Artemis launch finally happen again. We have our model behind us. We did a deep dive episode previously. So if you want a little bit more detail on actually what's happening, I will note the other last thing is they've actually changed the timing of Artemis 3 and 4. It looks like the goal now is mid-2027 for Artemis 3.
And then 2028 is when we're going to go to the moon. Some astute observers have sort of made the note that that is a very aggressive timeline. Yeah. There is a new head of NASA who's a private business type. Jared Isaacman? Yes. Yeah. And more of the Elon prototype versus a more research scientist type of prototype. Yeah. We will see what transpires. I do think it is an aggressive timeline. Yeah.
I think it's an aggressive timeline, but I think it's also correct. I think we need to be aggressive. I agree. Yeah. I agree with you. Yeah, I agree. Look, we all know who's coming for it. Yeah, we got to go. We got to go. We got to go now. Yes. We got to make it happen. Yes. If you've seen For All Mankind, you've literally seen this story before, and that's exactly what's happening is there's a clear geopolitical incentive by all of the spacefaring nations that the moon is – a strategic asset.
And whoever gets there first has free roam to establish, again, where is the helium-3, where potential ice and water deposits, all this stuff is going to matter. And if you're there first, it's hard for the other people to get into those locations. Exactly. It's a crude analogy, and I understand that it's not perfect because of the history of colonialism and everything. But if I may, I'd like to... compare this to the conquest of the new world. Yeah.
Right. Apparently the Norwegians were here in like 700 years to 400 years before Christopher Columbus came and the rest of the European nations. Right. But the Norwegians are not, we don't speak Norwegian. Right. In America. Right. We don't even speak Spanish because even though the British were like second to, I think probably fifth, They had the resources to build colonies, right, in this continent, in at least North America.
So if history has anything to say about this, we need to be there and we need to go there and we need to, like, actually stay and make a presence. Absolutely. And the last note to your point is as a part of the recent changes at NASA, they've canceled all the Mars related stuff. Yeah. And now we're saying instead of doing basically a moon base. Yeah. Is now the priority. Yeah. And I also get that. Right. It's like Mars. We've done a lot. And also there's still like rovers. Somehow JPL is so amazing.
Incredible. That like they plan for a two year rover and then it's still going like 15 years later. Right. Right. Like the wheels are broken and everything. I was like, no, I still got it. Still got it. So, yeah, I get it. And I'm looking forward to the next few years. I think it's going to be an amazing era for space exploration, for mankind in general. I can't wait for in about a few days if everything goes well and they do go to the moon. We're going to be able to look at the moon and cover it up with our thumb and be like, there are like four people.
Right there. There. Yeah. Like right now. Yeah. And they will be looking at the earth and they'll be able to cover it up with their thumb and be like, everyone except for four people are behind my thumb. It's, it's so. It's going to be incredible. It's like living in the 1970s. Like, imagine looking at the moon and being like, there's like four people there right now. Right. Fascinating. Always been a big lover of space. Yeah. That's something we both share. It's going to be amazing. So we'll see as things continue to progress. If you have. Yeah.
If you are a debunker about the moon landing, go watch our debunking video we already did and comment there. Moving on to our story number two. Anthropic accidentally leaked part of Claude Code's internal source code. So this, for those who may not know, Anthropic is one of the sort of primary... AI model creators along with Google's deep mind, open AI, Facebook's meta, whatever they're doing over there. There's a European player that no one ever talks about, but you know, these foundation models, the brains behind the chat based AI generative AI revolution we've seen have been crown jewels and is basically what the entire U S economy in terms of GDP growth in the last year or two is effectively based on.
So yeah, Whether you like it or not, our financial futures are kind of pegged to this stuff being successful in some way. But Anthropic just had a pretty embarrassing AI security issue, which has been a little bit overhyped in the headlines. Okay. But it's still somewhat significant. So Claude Code. So if we think about the structure, we have... Claude is the model, the actual brains that responds to you when you type into claude.com and get it back. This is an LLM? That's LLM, right. It's where all the model weights are.
It's where they do all this really expensive training to make these things super smart. They're basically growing the intellectual response that we get when they train these models. But Claude Code is simply a... interface and orchestrator for an architecture for you specifically initially for coding software coding but has now evolved into what they call co-work uh as well for people who do things like financial modeling uh deck creation other just business related functions they all live within this thing called the cloud code application you can either have it as a terminal or on the cloud desktop app okay now
Initially, this leak was reported on March 31st. And how it happened was... This is a little bit in the weeds, so I'll kind of keep it topical. There was a release of... a new version of Cloud Code, and on NPM, which is a package managing platform, there was a small addendum that was added by one of their internal developers that had a 60-megabyte JavaScript source map. And basically you were able to use that to reconstruct the entire TypeScript project.
So people were able to basically get 500,000 lines of the actual code base that is what Cloud Code is across roughly 1,900 files. And what was... fascinating to see was Korean developers, when they woke up, saw that this had happened, immediately pulled the package down, rebuilt Cloud Code, released it as basically a free open source version on GitHub, and it was the fastest growing repo on GitHub in history, reaching 50,000 stars in two hours, 100,000 in the first day, and it just exploded like wildfire.
I think part of that was because people were confusing the model For Claude Code. Oh, okay, I see. So they thought, oh, we just got the weights. Correct. Oh, my God, the weights are public. Correct. Okay, got it. Okay. Early on, people were like, oh, the weights went out. But it really is just that Claude Code. Right. They still have the crown jewels. And the irony of this story, there's a couple of things. It was basically an internal developer who screwed up with the release package. Yeah, someone's getting fired. They already got fired, and they tweeted about it. Oh, okay, great.
It's already on Twitter or X or whatever. Yeah. And so that went out. I think it's the speed in this space is astronomical. That is interesting. In a day, basically, developers are able to wake up, recreate it, not only in, they did it in like- Probably using cloud code. Exactly, exactly. And they recreated, I think, in JavaScript and then just built a version in Rust like immediately. And it's the ingenuity of developers now and because they have coding assistant tools, things can move very, very quickly.
Wow, that's incredible. And so the wrapper around the model leaked, not the core model. Not the core model, okay. But the other irony for Anthropic is when you look at the personalities of all the between Google, OpenAI, and Anthropic is the top three, arguably. Anthropic has been very heavy on the sort of AI safety. You know, we are focused on that piece and security and all this matters. So for them to have be the first ones to have such a catastrophic leak of something fundamental, it's not really good for the brand.
Mm-hmm. especially when OpenAI just finished closing their highest, the highest funding round, venture capital funding round. I think it was $122 billion in the history of funding ever, literally like a day or two before this happened. So the race is on. No, you're not just going to get, you know, clawed on your computer by default, but, you know, Cloud code is quite nice, and there's a lot of interesting insights in how they set up the harness around the model to make it practical that is going to continue to be used by developers outside of just using their tools.
Yeah. I mean, I'm sure by now there's probably like a dedicated team at OpenAI that is like, what's the best part of this that we can put into ours? I'm sure. I'm sure. And it's funny. There's like an undercover mode or something that was a part of the leaked code base that was specifically a tool to prevent this stuff from happening. Oh, hilarious. And clearly that did work. And that also got leaked. Jesus. Yeah. So tough day for Dario Amadei. Keep fighting the good fight.
AI safety is important. Operational security is also important. So that's our story number two. I'm going to take a brief pause for some show notes. If you love the show, everything you can do to support us is helpful, whether that's a like, a share, or a comment. Bringing us up at Journal Club, the billionaire algorithm, we have to continue to fight to get awareness and relevance. And all of that engagement really helps us get seen by more people. It continues to just be the two of us who are running this podcast. No advertisers at this moment in time.
Right. We're doing it for the love of the game. Yeah. And so if you want to support us financially, you can go to FFPPod.com backslash donate. And that is the place in which you can help support us in that way. But enough of all that nonsense. Let's move to our third story. So our third story is about a new science paper out in science. Yeah. that suggests that cats may be one of the best real-world models for human cancer and studying human cancer. It sounds a bit weird, but when you think about it, at the end of the day, it's all about animal models.
And we talk about this a lot on the podcast. The fact that when we do biomedical research, a lot of times we're looking for a model of a certain disease or a certain ailment, and we want to replicate that in some kind of... animal that we study, that we can study in great detail. A lot of time that has to do with rodents, right? Rodents are the bread and butter of a lot of biomedical labs. But rodents don't get cancer the same way that humans do, because they are not exposed naturally to the kind of carcinogens that humans are in day-to-day life.
So you have to induce cancer, you have to give them genetic mutations. Now, that... obviously has a lot of benefits, and there's a lot of incredible cancer research that goes on there. But there's parts that are missing. It turns out domesticated cats... live with human beings and are exposed to the same type of carcinogens that we are as human beings. And so the types of cancer that cats get, domesticated cats, are very similar to the types of cancers that humans get.
And so if we can study how domesticated cats get cancer, their genomes, their oncogenomes is what we should say, because this is specific to... the genomics represented in the tumors, in the cancer tumors, those are going to have all these different types of mutations that cause the cancer. If we can study that, that can then supplement a lot of the other research that's happening with cancer. And that is what this particular paper is doing. The paper was published in Science called The Oncogenome of the Domesticated Cat, which
It's from a bunch of collaborators, the University of Bern, from Cornell, from the Ontario Veterinary College, and from the Wellcome Sanger Institute. The researchers effectively sequenced about 500 individuals. 500 cat tumor and normal tissue pairs. So you get, you know, the tissue from the tumor and you get the tissue from the normal part of the cat. And so then you can compare what the mutations were. This is very similar to that story that we did of the guy in Australia who built an mRNA vaccine for his dog's cancer.
He did the same thing, right? He sequenced the normal genome and the genome of the tumor to find out how to target that tumor. Yeah. In this case, it's 500 pairs of genomes from 13 different tumor types. It's the first large-scale genomic atlas of cancer in domesticated cats. Obviously, this helps cats and vets more. understand how cats get cancer. But what I really like about this is this idea of the one medicine or one health approach.
That's the big idea. It's the fact that veterinary medicine and human medicine should not be treated as completely separate, siloed parts of medicine. Because cancer at the end of the day, if you look at a big picture, right, cancer is a genomic disease. And if we can study multiple species that get similar types of cancer, then we can study invariants that...
are unique to the cancer and the idea, the physics of cancer, let's say, from that first principle. And it has nothing to do with where the cancer is happening, whether it's in a cat or if it's in a human, right? It's a fundamental first principles understanding of what the cancer is doing. So they've identified 31 driver genes, and these are genes that drive cancer mutations later on. Okay, so you can have a gene that has a mutation and then that drives other types of mutations, right?
So that's what's called a driver gene. They identified 31 driver genes. There was this particular gene called TP53, and that one was the most frequently mutated gene showing up in about 33% of cat tumors. And that's a gene that's also going to show up in human cancers. So already we're getting a lot of these very familiar genetic motifs in both the human case and the cat case.
When it comes to carcinoma, one of the clearest parallels was this other gene called FBXW7. It was altered in more than 50% of the tumors that were studied in the cat. And again, that's another gene that's very, very prevalent in human beings. So if we can really start understanding how these cats get these... mutations, what are the carcinogenic sort of drivers of those kinds of mutations, then you can do something called comparative oncology, where if the cats and the humans get similar tumors, then perhaps the treatment and the biological insight can flow in both directions.
Right. We can use the human treatment to inform on how to treat a cat, and we can use the cat genomics to inform on the fundamentals of the human treatment. Cancer. So I thought it was a very cool study. I think one of the things from this story that immediately clicked in my head when we put in the show notes was... This idea that when we use, you know, we use animal models for research, you know, for a whole host of things. But this idea for cancer specifically that we basically induce, we artificially create cancer in mice and rats and other rodents we do in the lab.
And... It's a very simple idea. Domesticated cats live in the same environment as they're human. Yeah. And so their natural cancer is, from an environmental inducement perspective, is very closely mapped to our lived experience. And it makes very logical sense that... This is the whole nature-nurture thing, that the environment does create a better platform for study in terms of being able to take... Because we've talked about with ALS, I think it was a story where we talked about in animal models it doesn't translate.
Alzheimer's, animal models it doesn't translate. Consciousness, it doesn't translate. And so at least in a small... portion as it relates to naturally occurring cancers and oncology studies that that overlap to improving animal models made total sense yeah yeah that's cool to see in an actual study here yeah i thought that was very cool that was great we're gonna move on to our fourth story before we wrap up with project hail mary which is This headline in Reuters and TechCrunch, a major hacking tool has leaked online, putting millions of iPhones at risk.
And here's what you need to know. So we both have iPhones. Yes. I would like my iPhone not to be at risk. Yes, yeah, me too. And, you know, spyware, we've grown up in the generation where we've seen Facebook hacks and, you know, we've seen the rise of information warfare across nation states for a whole variety of geopolitical advantages. And so what this story is about is researchers have uncovered that two very powerful iPhone hacking toolkits, Dark Sword and Karuna, or at least parts of them, have leaked online in a public account.
GitHub repo, which just basically means anyone can download this. Oh, wow. And historically, these toolkits have been very lucrative for organizations to sell to nation states and intelligence community types and other non-state actors. Yeah. Who want to have targeted hacks of, you know... As long as you don't have your people leaking stuff on Signal, you want to be able to get to them. Yeah. So the software itself is actually not new.
The actual toolkits themselves. It's general availability for the regular public is what's new. Right. Yeah. This is just on GitHub. Right. Because normally something like this would be... tens of thousands, hundreds of thousands, millions of dollars to be able to, you know, have someone do on your behalf to get access. Because obviously Apple spends a ton of money and their whole brand identity is around privacy first. We're secure, local only, yada, yada, yada. So this idea that we're having government-grade exploits
Escape into the Wild, I think is the core idea for this story. It was first reported on March 18th. It should be noted that for Dark Sword, it specifically targets iOS 18.4 through 18.62. And what's interesting about iPhones... I'm going to check my iOS version right after this episode. Right. Right. What I will say is, you know, for most people, you're not doing anything that anyone is going to go through the effort to target as a practical matter. But also, one of the big differences between Android and iOS as ecosystems and the audiences is that...
Usually when there's new OS update, upwards of 80 to 90% of the iPhone install base just automatically accepts the auto update. And so the surface area of risk is relatively low, but the number looks to still be about 30%. 220 million to 270 million iPhones could still be exposed to this. So it's, it's non-zero. The downloads, you know, I are trying to be taken down from GitHub because obviously Apple has a lot of control there, but there's dozens of sort of sites in the, in Ukraine, Saudi Arabia, Turkey, Malaysia, et cetera, that,
You know, it's very prominent there. Wow. These are obviously areas where there's active conflict happening. And so, I mean, the story we just talked about where the astronauts on Artemis have iPhones. Yeah. They're not doing anything classified on there, hopefully. But it's, of course, technology that they want to preserve and protect. I was surprised that they just had off. They might not be off the shelf. Yeah. But I was surprised that it was an Apple product that they brought with them. It's curious. With Karuna, it has a wider spectrum.
It impacts iOS 13 through 17. So there's DarkSword, which is the first sort of hacking protocol. Yes. And then Karuna is the second one. Yes. Okay. And so DarkSword is more modern because it sort of targets iOS 18. 18, okay. Karuna is for the older iOS 13 to 17, which is much, much fewer phones, especially for high-value targets. It's less likely that they're going to be on these older versions. So... What does it allow the hacker to do? It can get your messages.
It can get your browsing data. It can get your location history and crypto-related information once the device has been compromised. And so, you know, it's not – the idea here is it's less about your iPhone is removable now, not only because most of us aren't doing anything that's actually worth the effort to – get it on there because you still have to install it and so a lot of times the way you get is when you get a website where it says oh you have something wrong click here to download a fix so you still need to get it onto the device but once it's onto the device it can then remotely execute a whole bunch of stuff um so like links and attachments and emails can be a point of a vector for attack attachments and i message etc etc so it is
Not ideal, but I think the bigger point here is it's similar to the drone issue that we've seen in warfare in the Middle East. Okay. Where the cost... is getting so low for things that used to be reserved for much more sophisticated shops. And it's not only happening on the battlefield. It's now also happening in the information war space, where now you're getting sort of what used to be a nation-state-only capability. Any Tom, Dick, or Harry can kind of get it off the shelf now.
Again, it is more of a signal about... The larger things to come as more of these tools, again, if Anthropic can't keep their model under wraps, if iPhone is still having exploits that are now getting to the public, it's a constant battlefield. So trying to do a little bit of separation between what is real versus what is hype. But I do think there is. This is a space to watch as spyware generally becomes more generally accessible on top of the fact that now there's all these coding assistants and coding agents.
So some high school kid who's clever, usually the best stuff comes from high school and early college kids. Right. You know, we've seen this cycle of... ransomware with hospitals where they lock down hospital system software until you pay in crypto. I don't think this is going away. And I've been surprised at how little has spilled into the public in all these proxy wars in terms of the information war component. But this leak, I think, is a signal that we're going to see more of this in the coming years. Yeah, that's pretty scary.
I mean, people, when they talk about... you know, leaks and like getting through cryptography. I think there's a lot of hype around, oh, quantum computing is going to, you know, change the game. And that's certainly true. Like if it works and we do have a fault tolerant, scalable quantum computer, it is going to break encryption, right? That's kind of an inevitability. But even before then, there's all these like cheap tricks. Yes. That... Probably we should be concerned about.
Yes. Make sure your admin password on any critical infrastructure software you guys have is not password or zero, zero, zero, zero. Yeah. So, again, just two leaked stories this week. You know, I think these are all getting a lot of attention. Yeah. And the coverage is not quite perfect. Well, I'm glad you covered it because I learned a lot there. Nice. We're going to move into Project Hail Mary. Yes. Our last note of the day in this rundown episode. You had seen it first. I saw it first, and then I told you to see it.
You told me to go see it. I was hyped about it because I love every space movie. All the early Rotten Tomatoes stuff, et cetera, was hot. Was hot, yeah. It was hot. So what did you think? You know what I mean? It was all right. It was all right. No, you know, I think that it was beautifully shot. Like, I love the visual language of the film. I thought the visual language was fantastic. The cinematography was fantastic. You know, I think the concepts were really interesting.
Like, also, spoiler alert, Warning. So if you haven't seen it yet, we are going to talk about the actual details. So like the Petrovillin concept and like the narrative tension of like what was happening and why it was important, I thought was great. I really had a hard time getting emotionally invested into the relationship between Ryan Gosling's character and Rocky. Oh, okay. I might have just been having an off day, and I was feeling a little bit like the Grinch, but I just couldn't get into it.
That's very interesting because I had the opposite experience. I was, like, super emotionally invested in Rocky the Alien, and I thought, like, as a film, I thought it was a masterpiece, honestly, because... I've seen a lot of films with probably a higher budget or a higher production value where I could not care less about the main characters. A good example is the Formula 1 F1 movie. At the end, I really couldn't care if Brad Pitt won.
Or if Lewis Hamilton won at the, at that last race, because like the way that they set up the character, I just wasn't, there was no arc. There was no like time to identify with him. I don't know why, but somehow the way that they, to me, at least, right. The way that they, they, they, had Rocky come up and like his struggles of like, he, like his crew died and all this other stuff. There's like, there was so much that I could identify with. I actually like, I think this was better than interstellar.
In terms of a movie, in terms of something that I could emotionally feel and be a part of. There are scenes in Interstellar that I emotionally resonated with. Like the fact when he was talking to his kids and he saw the kids just grow up in an hour. That was kind of crazy because obviously I know about general relativity and things like that. But seeing it on screen is pretty insane. Yeah. But yeah, I think in terms of that, like the movie and the characters, I thought it was amazing.
I don't agree. Okay. No, I'm really fascinated by that take. Yeah. Again, I have to probably watch it again. It just might have been not the right day for me. Yeah. But saying it's better than Interstellar is heresy. Yeah. To me, I will go ahead and say that. I think the science of Interstellar is cooler. For sure. But... I think in terms of a movie where I resonated with the characters and I was like emotionally locked in and I was just like, what's going to happen? Oh, my gosh. I think this one was like.
I think here's where maybe I will get a point of agreement with you, which is the one relationship that definitely was forced and didn't make sense or didn't have enough time to marinate was between Carl, the security guard driver, and Ryan Gosling's character. That was a bit rushed. They just immediately became buddy-buddy. And I was like, all right, that's a little too... It was a little too fast. A little too fast. So I think we can agree on that one. Yeah. I'll agree with you on that one. But let's continue. Yeah. So, okay. Well, anyways. So I want to cover some of the things that I found good about the science and some of the things that I was just like, that kind of doesn't make sense.
Again, I want to just say that... The movie itself, I still think, was just really well done. So let's start with the good. Yep. So I loved the way that the population dynamics, I thought, was very cool in that in Tau Ceti, The reason why that astrophage is not runaway exploding is because there's a predator. Predator prey dynamics is one of the first things that you learn about in differential equations and how there can be stable equilibria.
So I thought that was really cool. The idea that the acceleration of the spaceship was creating gravity and that once Ryan Gosling and his spaceship arrived at Tau Ceti... the spaceship stopped pushing yeah and then he was in zero g yeah i thought that was really cool like that switch because like when he when he first wakes up there's still gravity yes right and i was i was like what is going on how is there gravity in space if he's in space yes and then it switched off and then there was zero g i thought that was really cool that was a nice time
That was a really nice touch because I was like, ah, okay, now I see that. The details of Rocky the alien I thought were very cool in that even Rocky slept. This is something that we talked about in our episode about why jellyfish sleep, right? Jellyfish have a decentralized brain. They don't have a central nervous system. They don't have like... The reasons to sleep like we do, but yet they still sleep because it's a part of a DNA repair mechanism fundamentally at the end of the day and a cellular repair mechanism.
And so the fact that even Rocky had a sleep-wakes cycle. Because he's got some form of neurons, right? He's got some form of information capacity and information churning ability. And so to have even someone like Rocky sleep, I thought that was a really good touch. I think that is something that I would expect if we ever see aliens, intelligent aliens at least, that they would also sleep. His use of – so, you know, he uses echolocation to sense his environment.
So he doesn't have any eyes and he doesn't have any way to feel electromagnetic radiation because I think their species is evolved in a high pressure, very dark environment. So he used, I don't know if you saw, but he had like a little iPad that would make texture out of whatever Ryan Gosling was seeing. And he had a little crystal that would... take in electromagnetic rays and then transfer that into this like tactile ipad thing i thought that was pretty cool that was quite that was quite um because then he can see the ipad by bouncing sound waves yes off of it yes um that was clever yeah i i i thought that was clever i also like the fact that rocky had emotions i think there's a lot of like alien movies out there where like these aliens are like super emotionless and like don't
you know, that they're just super logical and things like that. I like the fact that Rocky had, like, emotions of sadness, happiness, like, missing his partner, things like that. Because if we look around in the animal kingdom, emotions are everywhere, right? In not just, I would say... um mammals but there there's evidence for emotion in birds and reptiles things like that and so it's kind of an evolutionary trick where you know there's some hormones or pheromones or chemicals in our brain that perhaps give us the illusion of emotion and it's like an evolutionary trick to get us to be attached to something or to an outcome or to an individual but i think that same kind of evolutionary trick would be there um
For an evolved system somewhere else. So I like that. Now, of course, these are like just soft, soft opinions. Totally. And I want to add two things, I think, to what I think the film did good. The sound design was out of this world. Yeah. I went to see it in IMAX. Oh, nice. Nice. The sound design is so good. Yeah. It is because it's space. Using silence well. Like a lot of times you hear sounds in space movies and it's like, oh, that's the sound from Jurassic Park mapped with a little bit or whatever, right?
It was very unique. They clearly did a lot of work there. The sound design was exceptional. And I just want to double click on how visually stunning it was. It was a beautiful, it was beautiful to watch and look at. Yeah. And, and just, and I think from an original screenplay design, screenplay perspective for a space movie. Yeah. I think they did a good job of not just repeating the same space movie again. Yes. I do. I do agree with that. The, Probably, I'm probably going to watch it again, but I really want to skip through, like, the part where Rocky, like, has to get out of his atmosphere and be in pain.
Because, like, that sound design was really well done because I was, like, feeling it because I had already attached, like, emotional weight to Rocky. And then I'm like, oh, like, this guy's dying. Like, so I was, I was, like, going through it. And I'm probably going to fast forward that part because I really did not enjoy watching that part. But the part before where they're like doing the fishing in the atmosphere, the score was absolutely incredible. Like, dude, everyone in the theater was locked in during that part.
Like everyone, you know that meme where they're like, you're like this and then you grab the controller. Everyone in the theater was like, all right, let's. That was probably one of the best intensity moments of the whole film. Yeah. And it was, that was quite nice. That was really, really cool. That was quite nice. So all of that I really liked. There are a few things about the science that I'm not sure I buy. Okay. Okay. And I want to cover this because a lot of the coverage that I'm seeing is that Andy Weyer, who's the author of the book, is, like, really nitty-gritty about the science, right?
So perhaps there are reasons because I haven't read the book. I did a bit of research to see, like, what the... what what the justification for these these points were i'm still not buying it so perhaps in the comments you can tell me if you disagree the first one has nothing to do with andy wire that just has to do with ryan gosling never having been in a molecular biology lab remember the centrifuge scene yeah have you seen all of the all of the scientists on tiktok and instagram no Everyone had like a gasp during the centrifuge scene because he puts two samples right next to each other on the centrifuge.
OK, I got a 15 minute lecture during my Princeton undergrad bio lab because I did that. And then like the centrifuge started doing this. And then the grad student who was my T.A. had to like unplug the centrifuge. And he was like, you're a physics major. Yeah. Like, are you... What are you doing? And then I thought about it for like five seconds. I'm like, yeah, that was really dumb. Because you got to balance. You got to put one on one side and one on the other side. And the thing is, it wasn't even like a...
The scene wasn't even off to the side. The camera focuses in on the centrifuge, and then he puts the two. So it's just full. The entire field of view is this centrifuge that I know is about to break, and I'm sure he only has one centrifuge up there. Right. So what? Now he's going to have to spin it manually or something, right? So that centrifuge definitely broke. Yeah. in the story. So that was pretty hilarious that everybody on TikTok was like going after that. The second thing was, so Ryan Gosling mentions that, or his character, mentions that Rocky's...
alien technology is like primitive in the sense that they don't know about relativity but they're really good at material science okay now they're so good at material science that they can literally build anything they want with the xenonite right right um which has xenon in some form i can totally buy that xenon has other allotropes that is not like gas, right? There's ways to manipulate xenon to be a material that we haven't really thought of. Condensed matters is a nascent field, even by today's standards.
So I can buy that. What I don't buy... is that an alien civilization is so advanced that they can literally 3D print Xenonite and things like that, but they don't know about relativity, okay? Because relativity to me is a very fundamental part of the fabric of the universe. I can forgive them for not understanding general relativity maybe, okay? But special relativity has everything to do with... just electromagnetism in general electricity and magnetism and i know for a fact that they know about electricity and magnetism for two reasons um well let's actually three reasons first um in the scene where ryan gosling is like trying to um in the scene where the the spaceship the earth hail mary spaceship is like
Turning around to create artificial gravity. There's antennas on Rocky's spaceship that are tracking. Those antennas are using light. Okay. They're using some kind of radar because that's what the only thing that goes in empty space is light. Right. Electromagnetic radiation. Right. So clearly he's using electromagnetic radiation in some rudimentary way, right? So you can say, oh, he just like, he knows it's like in some rudimentary way.
Fine. Second, he's literally 3D printing like entire spaceships worth of xenonite. Right. How are you manipulating that material if not with electromagnetic fields? Right. And if you're doing that... How do you not know about relativity? Because relativity is really a consequence of Maxwell's equations, right? If you know electricity and magnetism, you know if you're trying to manipulate charged particles, right, you know that when they move, they create a second type of force called magnetism.
And then you know that... The interplay between electricity and magnetism has to go at the speed of light. And then you can make logical sort of deductions about, well, what's the, like, how is this working, right? Like, it's really a logical argument what relativity is. And third, about the material science, right? There's two materials that I know for a fact Rocky knows exist in terms of elements. First, oxygen.
oxygen, he knows about oxygen because if you remember, there's a scene where Ryan Gosling goes through, goes to meet Rocky, and then Rocky's pointing at the little capsule that's a gift for him. That scene, by the way, was hilarious, where he kept pointing, and Ryan Gosling was like, oh, we want to keep doing it. Oh, you're pointing. He opens the thing, he sees two beads of eight and eight. Okay? And that's meant to be oxygen, because oxygen has eight protons and eight electrons and so an O2 molecule is 8 and 8 notice Rocky did not make 16 and 16 mm-hmm so Rocky knows about charge yeah okay yeah if he did 16 and 16 that means oh I just got it
I just weighed oxygen. That's good. Right? And it's like 16 things in here and 16 things in here. No, he did eight and eight, which means he's either referring to the protons or he's referring to the electrons, which means he knows about positive and negative charge. So he knows fundamentally about electricity and magnetism. knows about charge, Coulomb's Law, and they have some rudimentary form of understanding the mechanics of chemicals and chemical bonds. Yeah, yeah, yeah. Okay? So you can't tell me that he doesn't know about electricity and magnetism. That's very interesting. He has to know about light. He has to know that light is a manifestation of this thing.
That was, like, immediate to us in the 1800s when we figured out that The electricity and magnetism are governed by Maxwell's equations and the speed of light just pops out. The second thing, like along that vein, you're manipulating xenon. Xenon is a pretty heavy element. Right. Okay. Right. I think it's like 54 or something like that on the right hand side. When you're manipulating heavy elements, relativistic corrections become ever so important in chemistry.
Because the heavier an element is, the faster the electrons are moving. The electrons move close to the speed of light. And now the chemistry of that material has everything to do with the fact that as electrons move closer to the speed of light, they get heavier, which means the atom is going to get smaller, which means they're going to shield the... nucleus more which means the outer electrons are going to do weird things that's why gold is the color it is that's why mercury is liquid and that's i think if if he's manipulating xenonite at that fidelity that's there's no way that they don't understand relativity
It's a very compelling argument. I'm just saying. No, it's a very compelling argument. You can't be that good at material science and not know the fundamentals of our universe. Okay? It's not like physics is not something like, oh, I only need quantum mechanics. Well, no. In order to do that level of quantum mechanics, you need to understand, like, electron spin. Where does electron spin come from? That's from relativity. So I just don't buy that argument, that just because he uses echolocation— Means that, oh, he doesn't know about relativity. He doesn't know about light. We don't see the strong nuclear force, and yet we understand it pretty well.
Right. Right. Right. And also, it's one of those things where even if you're not, part of your point is like you would have stumbled upon it indirectly. Yeah. Even if you weren't trying to discover it directly. Yes. Yeah. Perhaps we had an easier time because we live in an environment where we can see things, where light travels long distances and blah, blah, blah. But I'm saying if they have the fidelity of material science ability that they do. In the movie, there's just no way that they haven't stumbled upon something as fundamental as relativity.
This is a hot take. You got to hear fresh off the griddle. I just don't think it makes sense. And if that's it, that's really good. That's it. That's it. Which is really good. And to be fair... I don't mind that this was part of the plot point, okay? Because I understand when it comes to, like, doing science fiction and art, you have to be able to, you know, take a little bit out. Because I think it's a big part of the plot that Rocky has extra fuel because he didn't take into account the fact that relativity makes his...
Right. Trip shorter. Right. So he had like over. Yes. Overcompensated and has a lot more fuel, which is why he can give some to Ryan Gosling. So I don't mind. I'm just pointing it out. Right. And I don't want to this. I don't think this takes away from how much I enjoyed the film. Right. Right. And how much I enjoyed the science behind the film. There's still so much that made me think as a scientist. And to be fair, this also made me think. a lot yeah so i very much enjoyed it i would recommend it to anyone and let me know if you guys think my take about relativity is correct i would be curious to see those comments we have gone long on this rundown episode thank you all for joining us for another week of fascinating stories we did not include
a segment of Are You Smarter Than a Scientist? Because we were so loaded with actual stories to cover. We will be sure to bring that back in the next rundown. I'm your host, Lester Nare, joined as always by my co-host and our resident PhD, Krishna Chowdhury. We will see you all next week.
Do pigeons navigate with their livers? A surprising study proposes that iron-rich immune cells help homing pigeons sense Earth's magnetic field. But other researchers argue that the mechanism does not hold up. Lester...
How can a weather satellite see Ramadan, Christmas and a blackout? Dr. Miguel Román joins Krishna Choudhary to explain NASA's Black Marble: nightly maps of the lights on Earth, from a single band on the VIIRS...
Can sound destroy a tumor without an incision? Dr. Zhen Xu joins Krishna Choudhary to explain histotripsy: focused ultrasound that creates tiny bubbles to mechanically break apart tissue, without relying on heat. From a...
What happens when AI moves beyond the chatbot and into the laboratory? From Claude-designed proteins that bind scorpion toxins to robots that assemble optics experiments, we examine what AI can do for science and what...
Can light restore aspects of sight? Can recalling a cold experience change an animal's metabolism? And what can ghost particles and mirror-image molecules tell us about the universe? Lester Nare and Krishna Choudhary...