mmm. he turned right back toward that kelp. oh, this is exciting. he's turning back toward it. filming with carlos is a pleasure, but i'm not the only one who wants to go out with him. scientists who are also interested in carlos�*s footage have taken an interest, including one local marine biologist who was with him when he filmed something he'd never seen before. | in the days both prior and the day| of, we had seen very large sharks. they had considerable girth on them. so it's a very wide—looking shark. so i'm like, "either that shark's | eating very well "or it might be i carrying some pups inside of it." female just...was acting erratically. she dove, disappeared. and guess what came up from underneath — this little bitty white, almost albino—looking white shark. i think the word albino was just shouting out of our mouths. i like, "albino, albino!" and then we start to zoom in on her. i'd never seen anything

like this before. - it was this white—looking shark. so it's called a white shark. you know, that's the species, . it's a white shark, but this shark was actually completely white in colour. - i was looking at the video, i'm like, "the fins — - "they're differently shaped, they're more rounded." - i'm like, "that's not... |that's not an older white shark, | "that's a very young white shark. "this could be a newborn." i'm like, "oh, my gosh, "oh, my goodness. - "this could be a newborn!" one of the greatest mysteries in great white shark science is reproduction. the holy grail of white shark research is the birth of a great white shark. and i believe that that is probably the closest we've ever been to the actual birth of a great white shark. phil believes the size and shape of the fins all point to a very young shark. he thinks the white layer could be residue from birth. if this is indeed a newborn white shark that still has this, - what we call embryonic layer, mucus—like substance - still on its body — - never been seen before. so this would be the first case that we've seen such a thing. _ who knows how old the shark actually is? -

is it minutes old, hours old, days old, weeks old? - | but nowhere else in literature has| such a finding ever been observed. but both men also say that this may be a skin condition. other experts accept this could be a historic discovery but that they need more evidence. everything that we know - about reproduction in white sharks comes from about ten females that were killed that were pregnant - at various stages of pregnancy. that's it. so we really don't know much, if anything, about white - shark reproduction. so i've seen the footage i of the shark, and it's really interesting and it's beautifully shot, of course, and - under great conditions. the problem that i have - as a scientist is it's what we call a sample size of one. sure, this could be a newborn shark, or it could be a shark— with a skin disease, _

or it could be a number of other things we haven't even thought of. but unfortunately, it's a sample size of one. | and a lot of people ask, "well, what would it take to convincel you that this is a pupping area?" and i would say, i would need to see babies coming out of a female. - but what everyone can agree on is that drones are playing a huge role in learning more about sharks, including how often they interact with humans. are there any moments that you filmed where you've felt nervous for the people in the water? i'd say yes. i'd be lying if i said i did not feel nervous sometimes. i it's not that we didn't know thati sharks were near people before — we've known that for a long time — we just didn't know how much. - the next question that. needs to be answered is, what does that mean? it's likely that drones will give us a tool to begin to answer them. i james clayton there.

and now for a change of gear. paul carter checks out a race with a difference. and despite the smooth overtaking, these racing manoeuvres aren't all that they seem. i'm at the las vegas motor speedway, and this is the indy autonomous challenge. and none of these cars have drivers — well, not human ones, at least. indy autonomous challenge is a research initiative where we bring universities from around the world and companies that are working in the autonomous vehicle space, and we try to advance the technology through the platform of motorsport by running fully autonomous race

cars with al drivers that are coded by top research university teams. the idea is to try to prove that autonomous technology can work at extremely high speeds. what you'll see is fully autonomous race cars racing each other, two at a time, passing each other at increasingly higher speeds until one car either crashes or gives up. and we can say that because there's no driver in it, so if it crashes, itjust costs some money. and, you know, the speeds that we've achieved in the past at this track have topped 180mph. this was a test day, where the various teams from top universities around the world were tweaking both the cars and their ai drivers to try and gain maximum advantage.

so what exactly are the challenges in trying to make an ai racing driver? well, there's multiple challenges, right? first is the perception, so getting all of the sensors, the lidar, the radar, the cameras to all work right. you have the communication technology, so how do you have really low latency communication? if the cars are moving that fast, you need to stay connected with them all around the track. and then there's the uncertainty of what the different ai drivers are going to do. so if a car is driving by itself, that's one challenge. if you put a second car or even a third car out there, now it has to predict what the other driver is going to do and that can be really difficult from a computational standpoint. everything that the opponent does, our car has to react to in real time the same way a race car driver would. as the car changes lanes, as it makes decisions to try to overtake, we have to predict and adapt

and move in order to make real—time decisions that could cost us the win or the loss. the challenges to making race cars race autonomously is exactly the same challenge as having a human do it. as you can tell, there's a very small percentage of people in the world who can actually race these cars, and imagine getting a team of software engineers together and to try to make that a reality in software that's making decisions in real time to compete against other software that's making decisions in real time. so it's not a trivial task. like regular autonomous vehicles, the race cars in the iac rely heavily on lidar to identify their surroundings. one of lidar�*s big advantages is that it can operate in low light. and to show this capability off, the main event of the track week was pitting the ai cars against each other completely in the dark. but as fun as watching self—driving racing cars hurtling

around in the dark is, this isn'tjust about entertainment. there is a serious side to all this. the main goal of this project is to advance understanding around vehicle autonomy, with the aim of ultimately making self—driving cars operate more safely at higher speeds. much of the industry has focused on low speed automation, kind of the urban, suburban robotaxi. we saw that — and in 2021, we came out with a new challenge to say, "what if we could prove that this technology could work at speeds, let's say, greater than ioomph?" and we think this is really important because, eventually, if we're going to have autonomous vehicles, we really want those vehicles to keep us safe at speeds that humans are not really comfortable driving. and once you get to 100 or 130mph, unless you're a professional race car driver, you really shouldn't be driving at that speed. but think of what it would mean for the world

to have cars moving people, moving freight at those kinds of speeds. the indy autonomous challenge is now expanding outside of the us, with races at iconic motorsports venues such as monza in italy and goodwood in the uk. autonomous driving technology continues to develop at speed, both on the road and on the racetrack. hello there. friday was a very warm day for april. temperatures widely reached the high teens and the warmest spots had temperatures a bit higher than that. 21.5 degrees celsius was recorded at both

stjames�*s park and northolt, both in greater london, making it the warmest day of the year so far. we have something of a change in the weather forecast, though, for saturday. cooler air is going to be arriving across both scotland and northern ireland behind this cold front. and that's significant because scotland also recorded its warmest day of 2024 as well on friday — 20.2 degrees celsius in fyvie castle. but across scotland and northern ireland, temperatures widely will be dropping by around five or six degrees celsius into saturday. so you will notice that significant change. right now the weather is quite quiet, though. we've got a few patches of high cloud around, one or two areas of mist and fog and a few showers coming in across northern areas with the winds strengthening slightly. temperatures as we start saturday morning around 8 to 11 degrees celsius. we're looking at a mild start to the day. now, a band of rain, a cold front will quickly swing its way into northern ireland and scotland. as that moves eastwards,

what will follow is blustery showers with some hailstorms mixed in the winds gusting up to 50 miles an hour further southwards. now a patch of rain for northern england and across wales, otherwise largely dry. much cooler air across scotland and northern ireland arrives through the afternoon, whereas across england and wales it's going to be the last of the really warm days. temperatures, i suspect, will reach around 21 or 22 in the warmest areas of eastern england, with colder air arriving across most parts of england and wales after dark. well, that takes us into sunday's forecast. and sunday is going to be a blustery day for scotland and northern ireland. frequent showers here, some of them with hail and thunder mixed in. there will be a few showers developing through the day across northern england and wales, but it may well stay largely dry across east anglia and the south of england. wherever you are, temperatures will be much closer to average, 15 degrees in london. so a big change from the low 20s that we'll have on saturday, it will feel cooler. on into the start of the new week, monday sees low pressure move in off

the atlantic with this cold front diving its way southwards across the country. and that's going to be bringing a spell of rain followed by showers, showery weather conditions around for tuesday and for wednesday. but the weather should become drier and a bit warmer towards the end of next week. bye for now.

live from washington. this is bbc news. the us says, the threat of an iranian attack on israel is real and viable — as president biden issues a clear warning to iran.

un agencies say parts of sudan may tip into famine if the civil war continues. we'll have rare pictures from the ground. and, joe biden seeks to boost his re—election campaign with a focus on abortion rights as trump hosts house speaker mike johnson to talk election integrity. hello, i'm sumi somaskanda. us presidentjoe biden is warning iran not to take military action against israel — as the us government says such an attack could happen �*imminently�*. iran vowed to retaliate after an apparent israeli air strike on its consulate in syria killed at least 13 people — including a senior military commander. that attack was condemned by the un's secretary—general, who said diplomatic and consular premises are off limits under international law. now, nearly two weeks later, american officials are telling cbs news an iranian strike on israel could come at any moment.

as concerns mount, joe biden is speaking