IRA FLATOW: This is Science Friday. I’m Ira Flatow. Sitting in with me this week is Sophie Bushwick, Senior News Editor at New Scientist. Hi, Sophie.

SOPHIE BUSHWICK: Hi, Ira.

IRA FLATOW: Later in the hour, the applications of AI in medicine. But first, this week, came news of an advance in animal cloning. Researchers in China have successfully cloned a rhesus monkey named Retro, that grew to adulthood and apparently lived normally for over two years. Here to fill us in on that is Timothy Revell, Deputy US Editor at New Scientist based here in New York. Welcome back, Tim.

TIMOTHY REVELL: Thanks for having me.

IRA FLATOW: Nice to have you. OK, tell us about this cloned monkey. What’s going on here?

TIMOTHY REVELL: Yeah, so there have been many attempts to clone rhesus monkeys over the years, but normally they result in very early deaths. And Retro appears to be the first cloned rhesus monkey that is completely healthy. It was actually born in July 2020, but we’re just hearing about him now. So he’s actually more than three.

And the thing with this clone is that it’s slightly different cloning to the normal type of cloning you’d think of in terms of Dolly, the sheep, from the 1990s.

IRA FLATOW: Right.

TIMOTHY REVELL: And that’s that rather than using adult cells, fetal cells were the key thing at the beginning. So that means you couldn’t use this technique to just take some cells from you and create a second Ira. Instead–

IRA FLATOW: Who would want that to begin with?

TIMOTHY REVELL: Instead, you would have had to imagine that right at the beginning of your life.

IRA FLATOW: And what did that solve? Was there a specific problem that that solved?

TIMOTHY REVELL: Yeah, the difficult thing, particularly with primates and with humans, if we’re ever to try to clone them, is that throughout your life, you get lots of these genetic markers that change how your DNA behaves. And they work well for as you’re older, but they’re not the right markers for if you want to create a completely new you from the beginning.

IRA FLATOW: Mhm.

TIMOTHY REVELL: And so what– by taking fetal cells, rather than adult cells, it doesn’t have any of those genetic markers on them. And therefore, you can create a clone that way.

IRA FLATOW: Very clever.

TIMOTHY REVELL: Yeah, very clever, indeed.

IRA FLATOW: But is this going to lead to more cloned primates, do you think?

TIMOTHY REVELL: I think it’s possible that this method might be used for one specific use case, and that’s for if you want to have a scientific study where you have lots of primates that have all got the exact same genetics and you wanted to test different medications on them. But primate research is very controversial, and so is cloning. And so I don’t think we’re going to see a quick explosion of this technique.

IRA FLATOW: Or people.

TIMOTHY REVELL: Or people. I honestly think that it was so difficult to produce this–

IRA FLATOW: It was, yeah.

TIMOTHY REVELL: –rhesus monkey, that it seems unlikely to me– it’s illegal in most countries to try a human clone. As far as we know, nobody’s actually tried that.

IRA FLATOW: Yeah. OK, let’s move on to news as we speak about a Japanese moon lander called SLIM. What is this?

TIMOTHY REVELL: Yeah, so this is an amazing moon lander. SLIM stands for Smart Lander for Investigating the Moon. And since December, it’s been orbiting the moon. And within the last few hours, it actually touched down on the moon. And the thing with SLIM is that it was built to test this technology called smart eyes, which is all about how accurately could we possibly land on the moon.

And so normally, the area you’re looking to land is tens of kilometers wide. But the hope– and this is still currently being confirmed as we speak by the Japanese Space Agency– the hope was that it would be able to land in an area as small as 100 meters.

IRA FLATOW: Wow.

TIMOTHY REVELL: Yeah, really–

IRA FLATOW: Wow.

TIMOTHY REVELL: –way better than anything else.

IRA FLATOW: They have some proprietary method of doing this?

TIMOTHY REVELL: It’s just like the technology that exists today. So things like artificial intelligence. So as it descends, it’s using image recognition to spot craters and so course-correct automatically as it goes down.

IRA FLATOW: That is really cool.

SOPHIE BUSHWICK: That’s pretty cool.

TIMOTHY REVELL: It’s really amazing. So even just landing has been– is a really good achievement. The thing that is a slight disappointment is it seems that the lander’s solar panels aren’t working.

IRA FLATOW: Oh!

TIMOTHY REVELL: And if that’s the case, it’s only got a few hours battery life from now. So JAXA, the Japanese space agency, they’re currently rushing to try to fix it. But also you make the most of the few hours that they might have.

IRA FLATOW: I get it. OK, let’s move on to a story that you have about human life expectancy, a gap between men and women.

TIMOTHY REVELL: Yeah, this is a really cool study that’s looked at mortality data from 194 countries over a 20-year period. And it’s about how life expectancy is changing, and how it’s changing between men and women. So over that 20-year period, nearly every country life expectancy has increased. And throughout that period, also, the gap between how long men and women has also decreased.

IRA FLATOW: Right.

TIMOTHY REVELL: So typically, women live a little bit longer than men. And in the richest countries that they group together in this research, that gap has been closing a small amount, from 4.85 years at the beginning of the study to 4.77 at the end, which is only small. But the team predict that, by 2030, the gap will be down to 3.4 years.

IRA FLATOW: That’s amazing. But I know–

SOPHIE BUSHWICK: Do they know why?

IRA FLATOW: Yeah, why?

TIMOTHY REVELL: Good question. Yeah. So part of the reason is down to– so life expectancy is generally increased because medicine has improved. And one of the areas where we are seeing quite a big impact is on diseases such as those related to smoking and alcohol, both the medicine– the treatments available but also awareness campaigns. And those typically hit men harder. More men were affected. And so as that’s reduced, the gap has decreased.

IRA FLATOW: And this is a global thing.

TIMOTHY REVELL: This is a global thing. Yeah, there are only a handful of countries where this pattern is not what we’re seeing.

IRA FLATOW: Looking at numbers– speaking of numbers– you have a story about AI doing tricky math problems. I could have used that years ago.

TIMOTHY REVELL: [LAUGHS] Yeah, me, too. I really love this story. This one’s about– you’ve probably heard of the International Mathematical Olympiad that takes place–

IRA FLATOW: Oh, yeah.

TIMOTHY REVELL: Yeah, it’s basically the big math competition for high-schoolers, and it pits the mathematical wits against each other. And traditionally, AI is terrible at mathematics. So GPT 4, for example, OpenAI’s really famous AI, scores zero on math Olympiad questions. It just can’t do them. And even specialized AIs built to try and solve this problem have not been particularly good at it.

Well, now what’s changed is that Google DeepMind have had a go. And they’ve built this AI called Alpha Geometry. And on a test of 30 Olympiad questions, it scored 25. And by comparison, an Olympiad gold medalist, which is the best of the best, is expected to score about 25.9. So it’s hot on the heels of the smartest high school math Olympiads.

IRA FLATOW: Yeah, I remember doing geometry. I really liked it because you had to do proofs of things. And that’s what they must have had to do, right?

TIMOTHY REVELL: Yeah.

IRA FLATOW: The AI?

TIMOTHY REVELL: So that’s exactly right. And one of the ways that they improved their AI to make it possible is that it’s actually built in two parts. The first part is a bit like ChatGPT. It’s good on vagueness and language and understanding the problem, but it’s not the one that’s allowed to solve the problems. Instead, it feeds it to another part of the AI that has to use rigorous mathematical logic, which is what gets it to an actual proof and an answer rather than the fake answers that you tend to get with ChatGPT.

IRA FLATOW: [LAUGHS] Imagine the homework it had to do.

TIMOTHY REVELL: Yeah. It was trained on hundreds of millions of examples.

IRA FLATOW: Oh, is that right?

TIMOTHY REVELL: Hundreds of millions, yeah. So it did a lot of homework before it became good enough.

[LAUGHTER]

IRA FLATOW: Better it than me. Let’s move off the planet for a minute. There’s a mysterious patch on Mars. I mean, there’s always something mysterious on Mars.

[LAUGHTER]

But this appeared to be a giant lump of ice?

TIMOTHY REVELL: Yeah, it’s amazing. So it’s this huge deposit around the equator of Mars, called the Medusae Fossae formation. And we’ve known about this for 15, 20 years, but not been exactly sure what it is. It turns out now, according to new data, that it’s actually this patch of water ice. And there’s so much water ice there that if it melted, it could cover the entire surface of Mars in 6.5 feet deep of water.

IRA FLATOW: I’m just trying to absorb that. That’s– so it’s a giant patch of ice that if it melt– wow.

TIMOTHY REVELL: Yeah, it’s like a ring–

IRA FLATOW: That’s unbelievable.

TIMOTHY REVELL: –a ring around the equator. It’s amazing.

SOPHIE BUSHWICK: And I’m just picturing that amazing planet-wide swimming pool. [LAUGHS]

TIMOTHY REVELL: Yeah, it would give a very different vibe to the red planet if it–

[LAUGHTER]

–it would be the blue planet, wouldn’t it?

IRA FLATOW: So it must tell us something about past that this water came from somewhere, right?

TIMOTHY REVELL: Yeah, so the current conditions on Mars would not allow this sort of water to form. So instead, the suggestion is perhaps– we know that Mars over its lifetime has tilted many times and swung back and forth. And that perhaps in its past, this water ice formed when the equator was pointing further away from the sun.

IRA FLATOW: Let’s stay, because it is Science Friday, in space news, which we love. There’s some black hole news this week.

TIMOTHY REVELL: Yeah, the James Webb Space Telescope, or the James Welly Space Telly, as my space colleague likes to call it–

[LAUGHTER]

IRA FLATOW: I like that.

TIMOTHY REVELL: Yeah, it spotted this black hole that is the oldest and most distant black hole we’ve ever seen. The black hole is six million times as massive as the sun. And it’s located in this galaxy named GNZ11, and that’s about 13.4 billion light years away. But the light coming from that galaxy is just 400 million years after the Big Bang. So it’s really, really young.

IRA FLATOW: Wow, that’s just like it’s in its infancy, right? Right after it formed. Does it tell us anything about black holes or is it just the oldest one?

TIMOTHY REVELL: Yeah, it’s the oldest one. But it’s very strange for a black hole to be that big, that early in the universe’s history. So it suggests that there might be something wrong with our understanding of how black holes form.

IRA FLATOW: You think?

[LAUGHTER]

We keep hearing new stories about black holes being mysterious. Well, our laws of physics are not up to snuff anymore.

TIMOTHY REVELL: Yeah, they’re not up to snuff at all.

IRA FLATOW: [LAUGHS] How– let’s talk about global warming is turning ibex nocturnal, is that right?

TIMOTHY REVELL: That’s correct. Yeah, so this is specifically ibex– Alpine ibex in the European Alps. And as global temperatures warm up, the animals that are most affected by that tend to be the ones that are in colder climates, such as these Alpine ibex.

And a team looked at their patterns of behavior over a 15-year period and found that the ibex are– when it’s hotter during the day, that they become much more active at night. But the thing with that is that at night, wolves are much more active, too, and they’re particularly partial to the taste of an Alpine ibex.

SOPHIE BUSHWICK: Oh, that’s not good news for them.

TIMOTHY REVELL: Not good news for them.

IRA FLATOW: No. Finally, we have this story that’s both materials and topology– the smallest knot.

TIMOTHY REVELL: Yeah, smallest and tightest.

IRA FLATOW: How small are we talking about?

TIMOTHY REVELL: Just 54 atoms form this knot. And it’s formed into a sort of trefoil knot shape, so three intersecting parts. And, yeah, it’s pretty amazing. It’s got– it’s partly made of gold. And it was almost made serendipitously because it was just mixing certain atoms together resulted in this knot.

IRA FLATOW: Hmm. And so it formed spontaneously when they mixed the atoms together?

TIMOTHY REVELL: Yeah, that’s right. So they mixed together a liquid containing gold atoms linked by carbon rings and then phosphorus atoms linked by a different assortment of carbon rings. And then they don’t know why it forms, but it does. And then they ended up with this amazing knot.

IRA FLATOW: And they were trying to do something else probably.

TIMOTHY REVELL: Well, they were hoping to form knots. The thing is they just don’t know why they form this way. But the hope is that if they keep doing this, eventually we might figure out why the knots form. And that could help us to make more interesting materials, but also to better understand biology, as proteins often form these strange types of knots.

IRA FLATOW: Or not.

TIMOTHY REVELL: Yeah. Or not.

[LAUGHTER]

IRA FLATOW: Thank you very much, Timothy. It’s great stuff.

TIMOTHY REVELL: Thanks for having me.

IRA FLATOW: Timothy Revell, Deputy US Editor at New Scientist based in New York. I understand you’re heading back to London shortly. You’re going to be leaving us.

TIMOTHY REVELL: Yeah. In March, I will be back in the UK.

IRA FLATOW: Well, good luck to you. And thank you for all the work you’ve been doing with us. We’ll still be in touch.

TIMOTHY REVELL: Yeah, I hope so. I can dial in from there.

Copyright © 2023 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/