Dan Snow's History Hit - How AI is Safeguarding Maritime Heritage

Episode Date: March 26, 2020

There are more historic artefacts on our ocean floor than there are in every museum in the world put together. Over thousands of years ships carrying every conceivable cargo have sunk in the rivers and oceans of the world. Protecting them is an enormous challenge. Thankfully there are heroes out there who are taking on that challenge. In this episode I was lucky enough to talk to maritime archaeologist Jessica Berry, CEO of MAST (Maritime Archaeological Sea Trust) and her colleague Nick Wise. Nick is CEO of OceanMind. OceanMind is a not for profit which specialises in using the latest technology, unleashing the full potential of the latest AI from Microsoft. Together their two organisations have now set up the Maritime Observatory. This will protect underwater heritage from illegal looting - people going down, and ripping off bits of wrecks. So what Microsoft AI, and OceanMind's technology, allows MAST to keep an eye on ships behaving suspiciously on the surface in areas of important maritime wreck and heritage sites. All this means that wrecks can be looked after, war graves can be respected, in just the same way we should be looking after our vital fisheries and our maritime protected areas. To learn more go to Microsoft.com/ai For ad free versions of our entire podcast archive and hundreds of hours of history documentaries, interviews and films, including our new in depth documentary about the bombing war featuring James Holland and other historians, please signup to www.HistoryHit.TV We have got a special offer on at the moment- use code 'pod3' for a month free and the first THREE months for just £/€/$1 per month.For information regarding your data privacy, visit acast.com/privacy

Hi everybody, welcome to History Hit. I'm very excited about this episode because this episode we get to talk about the sea, the big blue wobbly thing that so many of us can see if we look out of our windows now. And for those of you who can't, I tell you it's looking beautiful. And this is all about maritime archaeology, which as you'll know, I have a little soft spot for. And even better than that, it's about how cutting edge technology, some of the most exciting technology you've heard about, is changing and aiding our quest to understand, to protect, and to learn more about our seabeds and what's on them. This episode is brought to you in association with Microsoft AI, who are basically allowing the rest of us to unleash our potential for innovation.

I'm thrilled. You're going to love this episode if I don't say so myself. As you know, I am strangely passionate about maritime history. I'm never happier. Well, my happiest, of course, is enjoying domestic bliss with my wife and children. But apart from that, I'm never happy professionally, let's say, than when I'm out in the intertidal zone, looking at shipwrecks or beneath the water diving. And I come from an island in the North Atlantic. The powerful tides swirl around these shores. There are shallows, there are reefs, there are shoals, there are rocks, there are great chunks of limestone sticking out of it. The gigantic weather systems that blast our rocky shores with near constant

southwesterly winds have been doing so for millennia. And as a result, these shallow seas around the coast of where I'm from, a little place called Britain, are full of shipwrecks. I mean, absolutely chock full of shipwrecks. You could probably paddle. You could walk east to west along the south coast of England. Well, if you had extremely long legs and a pair of waders on shipwrecks. I urge you not to do that, though, because we need to protect them. So I'm glad to say on this episode of the podcast, we're going to have a big look at underwater archaeology and the technology that's protecting those vital sites. We've got a combination. We've got a mashup. We've got Jessica Berry. She's a maritime archaeologist.

Brilliant author. Part of the team excavating the Swash Channel wreck, a 17th century armed merchant ship off Poole, which is very near where I live. So I love that one. She's also at SCOTED HMS Invincible on the South Coast. She's CEO of MAST, Maritime Archaeological Sea Trust. It's a champion for underwater archaeology

and all maritime heritage, really. And she's now joined Forster with the very brilliant Nick Wise, who you'll also be hearing from. He is the CEO of Ocean Mind. Ocean Mind is a not-for-profit which specialises in using this latest technology that's been provided for them by Microsoft AI. Now, you've all heard of Microsoft. They are obviously transforming the way we live and work in every other field. But what is exciting to me is I didn't realise what they're enabling within the maritime heritage space. Ocean Mind is building on Microsoft technology to help governments and the private sector do things like monitor illegal and unregulated fishing.

And so MAST and Ocean Mind have now hooked up to create something called the Maritime Observatory. And this is really cool because their job is to protect underwater heritage and prevent looting on these shipwrecks. People going down, ripping bits off. We don't want that. And so this technology is allowing OceanMind to keep an eye on the shipwrecks, to keep an eye on the surface of the sea. So any suspicious ships coming, going, hovering around, looking a bit dodgy, the relevant authorities can be notified and they can send out ships to deal with it and this is so important because it means the wrecks can be looked after war graves can be respected really in just the same way that we're looking after hopefully vital fisheries and maritime protected areas so in this podcast talking to nick and jessica i'm going to be learning about

the maritime observatory learning about what the future holds, and all this wonderful potential for harnessing technology to transform the way we look after our maritime heritage. This podcast has been made possible by Microsoft, so thank you to them. I love learning about this tech, and I really believe that tech, married to existing organizations and proper enforcement, will transform our behavior on the oceans. And it means we can be more sustainable, we'll be more responsible, and we can learn more from the heritage. Because as you'll hear Jessica reminding us,

there are a lot more historical artefacts on the floor of the oceans than in every single museum on the land. It's all down there, people. We're going to look after it. If you are interested in any of the issues raised in this podcast, please go to microsoft.com slash ai to learn more guys thank you very much for coming on the show this is very very exciting indeed because i love maritime archaeology and so it's great honor for me to have you here i'm jessica one of the legends of the field. Give me a sense of the scale

of the problem of trying to catalogue and then let alone protect the wrecks, the sites that lie off our shores. Well, if you think about Britain as having been an island for quite a long time, we've got more wrecks off our shore than any other country on Earth. That's just the starting point. And then if you think about the fact that the Royal Navy from the time of the 17th century was the greatest in Europe in terms of its exploration, the age of exploration, the age of sail and so on and so forth. There are more Royal Navy wrecks than I think there are from any other country, similar military wrecks. And we did a bunch of research on Royal Navy wrecks. I compiled something called the Royal Navy Lost List, and we are up to just under 5,000 around the world since 1600. So that's an awful

lot of wrecks, not to say we know where they all are. And the Royal Navy weren't quite as good in some circumstances of knowing where they were when they were going down than the Merchant Navy and that's just a spat of their own. That means that especially in First and Second World War when the numbers obviously peak there are an awful lot of metal wrecks and the oil fired ones are the ones that we are also particularly interested for one reason because of their potentially leaky status and also they have naturally because they're warships unstable ordnance so that's something we've been looking at as part of the maritime observatory which

encompasses heritage from what we think of as real archaeology so wooden ships from HMS Victory Gloucester and other very important warships and sites of archaeological interest to the ones that are not only archaeologically interesting but also can cause untold damage environmentally too and also present a gold mine as it were to looters who see the scrap metal as of an extraordinary value that's the sense of the maritime observatory is to have eyes on thousands thousands of sites yeah i mean you've mentioned my favorites there gloucester victory can't wait i mean amazing but nick tell me what is the observatory and why can we now why are we now this ambitious how could we even dream that we can properly survey these thousands of

sites that we've just been hearing about? So the observatory really is two things. One part of it is the expertise of understanding wrecks and the sort of life cycle of wrecks and how people are interested in wrecks and what sorts of activities take place over wrecks. And the other then is technology and other capabilities that allow you to look into those activities. And so really the game changer is satellites, where before you'd have to go somewhere to see what was happening there, particularly the very remote RECs. Now you can use satellites to get imagery, which is not just pictures, but also radar and other types of sensor like infrared,

to work out what's happening over these wrecks. And you can also start to use things like satellite radar to see oil slicks and signs of potential leakage over the top of wrecks. So you can use that as an early warning system for when there might be a more catastrophic leak. So really, satellites combine that with artificial intelligence to start to understand from all of that massive data, because it's a massive data, what the relevant behaviours are that you're looking for

that indicate a potential set of activities over a site that shouldn't be there, whether that's the leaking from the wreck or whether that's people trying to take things from the wreck that they shouldn't. And then you can start To build an infrastructure around that to get the experts informed as to what's happening so that we can then work out how to take action But the AI thing is so fascinating because the dream has always been That we don't need human beings looking through I mean our friends of mine have worked on like crowds Anonymizing and crowdsourcing satellite data to you to survey sites all over the world or even look for new sites. But now you're saying that the AI can just go, it's always on. So it's just

all the data that's coming up, like ship movements, oil slicks, whatever it might be, the AI, you just turn it on and then you go to the panel. I would say it's not quite that good, as you might imagine. So the key about the AI is it helps filter the data down. So there's not a hope that any human being is going to review all of this data. And therefore, we would never be able to see some of the things that we can see. So really, the AI is there so that we can use the experts to train the AI to recognize the things they're looking for. And therefore, the AI takes a first pass through the data. And it tries to find some of the indicators

that the human experts would have been looking for. And where there are absolutely no such indicators, then we can conveniently say, we don't need to look at that data because it's clearly not relevant to the problem space. So we're actually cutting the data set down significantly. And then the humans look at the relevant data set, massively reducing their workload. And then, of course, the AI can

start to prioritise things. So there are certain indicators that are much more relevant than others. So if we're talking about environmental threats, then an obvious oil slick is something that needs immediate attention. If you're talking about illegal salvage, then a vessel that shouldn't be there hanging around on the site for several days, that needs immediate attention. But there are other more vague indicators that allow us to then work out whether we need to get someone's eyes on that piece of data to work out what's happening. So is this project now up and running?

As we sit here, are we crunching through this data? So we are crunching through the data and we're already crunching through the data because our primary work at Ocean Mind came from the fisheries world. So we're crunching this data looking for things like illegal fishing or human trafficking on fishing vessels. And when we met Jessica and her team, it became obvious that the same sorts of work would be relevant to the salvage world because we're looking for specific patterns of behavior and that's exactly what we do in fishing we're looking for particular patterns of behavior and then looking at the rules and regulations around that to work out whether that behavior

is good or needs investigation this pretty much is the same kind of thing on the same kinds of data. So it was a logical step to then move from that world into starting to look at the Rex world. And so yes, we're crunching lots of data, but not necessarily over every site at the moment, because there is a huge amount of data. And so that costs to process. And so we're picking those sites that in the first instance are best value and then working out who to work with to help fund moving forwards. Jessica are you identifying the sites for them? Yes we started off with what we've called a proof of concept and we used data over it was three years wasn't it? Yeah. And we fed into the computer about a dozen important sites

around the world that are you allowed to tell me what they are i'd rather not okay it's one of them the enormous munition ship that might blow up and take out north no it isn't okay because that's perfectly well looked after okay the um ss montgomery montgomery no but there are about a dozen sites around the world from North Cape to East Asia to our own waters. And not only the sites, but we're also particularly interested in some known vessels of interest. Ah, looters. Yes. And, you know, there are a few that are quite well known to this government and other European governments. So we followed them using legacy data to various sites around the world. What's legacy data mean?

Old, archival stuff. Not only is there data you can gather now, there's a huge mountain of data in the past that was just gathered. I think it goes back up to three years AIS, using AIS. AIS, which is the sort of tracking beacon on all ships, commercial ships. Over a certain tonnage, yeah. So we discovered things that we didn't know before about what they'd been doing in sites that we didn't know had been found.

And then we're able to tell the relevant authorities and the relevant governments. This is the best news I've ever heard in my life. So you're saying, we're not going to mention names, but for example, very important Second World War casualties in East Asia, which have been the focus of looting and scrap metal merchants. We can now protect those more effectively. So the work that you have done has led to actual real world outcomes

on the surface, has it? It has, with certain sites. I mean, the one you're mentioning, I think HMS Prince of Wales or HMS Exeter, I think you have one of the two you think of. That has already come to pass before we came along. But obviously there's still significant interest in that area, particularly since some of those vessels are increasingly vulnerable because their hulls are getting thinner and thinner. It's not helped by the use of explosives by people who've been working on them. What do most of these sites sidetrack?

What are these looters interested in with these vessels? Primarily, it's scrap metal, and there is also a school of thought that pre-nuclear metals, that is to say before Hiroshima and Nagasaki, have more value and are used in medical instruments and that kind of thing. Is that an urban myth myth or is that true? I don't think it's an urban myth, but given the scale of the metal and most of it, I don't think

could be transferred to those very, very fine instruments. But it is maybe a small byproduct, I would say, rather than the whole thing. And there are some horrendous images, which you can just find on Google of ships and submarines raised wholesale from the seabed and sat on the flatbed of chinese vessels and um obviously because submarines in themselves by their nature are tombs you get the um bodies of the dead that come up in part and one instance where we know that they were put into ziplock bags and chucked into shallow graves so it's it's pretty horrendous stuff but but so not the for example prince wells but there are now sites where this project is in action and you are you have there have been real

world consequences yes that must be enormously rewarding well it is extremely rewarding and now what we need are governments and subscribers are different subscribers to sign up and join so we're talking to a couple of european governments at the moment and our own and moving forward and then the key is that um the only people who have the authority to do something about it are the responsible governments. So they have to be prepared to take action if we find something. And that really is the key for all of these compliance-related issues. You need a relevant authority to decide to take action on something when you provide the information. And some are relatively good, and some are maybe not so responsive at these things.

Why do you think governments, is it corruption? Is it they still don't really understand the importance of the maritime heritage environment? Why do you think there's a reluctance? Is it just because it's new technology and you're blowing their minds? Well, I would say, and it's common in other industries as well, like the fishing industry, that it is not high on government's agenda

and therefore not highly resourced. And the training that people might have is not necessarily up to the right level. Certainly when you're talking about developing nations or lower middle income nations where they have responsibilities. In Western nations, you find that issues like this and indeed fisheries and other sort of less high profile issues don't receive the attention and the budgets and the resources that they deserve. And so part of the problem is finding someone who can take action. What's our experience in the past is it's quite easy to embarrass governments or well, this government, into action and something must be done and there's a splash in front of Daily Mail and then time passes and things continue, which is why we started the Maritime Observatory

because we just can't keep going from these start-stop situations. It feels like you're sort of taking matters into your own hands. That's pretty much what we've done isn't it? To a degree but not with the intention that we're going to do it on anybody's behalf so to speak. We still need engagement but what we want to do is apply sort of the large scale economies of scale that you can get from this data processing to make it more sort of convenient and straightforward for governments to get involved and be able to make use of the capability.

So it's more sort of lowering the barrier to entry is what we're hoping to achieve. What else might this technology offer us? I mean, it's a bit depressing, but can looters actually show you where some previously unidentified wreck sites might be? I mean, ballot station might be too late, but are there other useful things coming from this project?

Well, we've suddenly established interesting patterns of life, which are slightly different to the way illegal fishing occurs, which is why there's two separate teams doing different bits of analysis. But we've noticed that certain vessels working together, which we had no idea before. So we can start tracking people with more knowledge in the future because we've worked out what their MOs are. And is this ever going to be a sort of surveying tool?

Is it ever going to allow you to learn new things about the ocean and what lies under there? Or is this always going to be a is policing the right word well i think what we're looking at is not prevention but deterrence and as we develop and we have eyes further afield the ai develops and we develop with it and we start working out patterns of behavior that we weren't aware of so we can start predicting things that we currently are unable to do. And that's just by building up our knowledge and the computer system's knowledge. Would that be correct?

Yeah, and certainly we consider ourselves as being in the business of intelligence. So we're producing information that others will largely act upon, which includes the policing authorities and organisations like Interpol for some of these other crimes. But we're not going to be taking action ourselves. We're not going to be like a vigilante force, as you might see in other areas, because that's not the effective long-term way of doing it. And you might be able to do a bit of that, but you can't keep that going for a long time. It's really the responsibility of governments. And what we be able to do a bit of that, but you can't keep that going for a long time. It's really the responsibility of governments. And what we would like to see is that those governments step up to that responsibility,

given that there is a more broadly available capability. But as Jessica says, over time, we'll learn more things. And so it may well be that we start to understand the things about the ocean that we didn't previously realise. Particularly, we might be able to, it's feasible that we could spot new sites or more accurately pinpoint sites where currently it's in this body of water, we might be able to work out more clearly where off the back of some vessel surveys and so forth we'll be able to see what they were doing and then we can start to understand why they were doing it and where it is so it's all a bit hand wavy and vague but it looks like in the future

that might be possible but it's not now so what else i mean so i mean i'm so excited by the opportunities for technology to help cultural protection around the world both on land and sea i mean what else are you you excited about over the next few years? I think the main thing is, it's really the merging of technology with expertise that makes the difference. And it's absolutely not the case that the technology in any way supplants expertise. We need to understand first what we're looking for before we can start to train computer systems to help us look for those things. Because this is a human activity we're trying to understand. And the problems surrounding that process and enforcement and all of that are all human problems. So we're trying to inform those human problems. I think the technology that forms part of the observatory is being used in various

different areas at the moment. I think what we'll see over the coming years is a convergence of those capabilities. So where at the moment we're doing illegal fishing as a thing, or we're doing the illegal salvage as a thing, or we're doing human trafficking in fisheries related things as a thing. These things will start to merge together and we'll start to see a higher level picture, higher level understanding of the human behavior on the oceans and start to see the patterns of life in more detail and start to understand them at a higher level because almost always these patterns of life relate to some form

of economic behavior and there's various social and economic drivers that cause people to do things and the fact that the ocean is, although it's highly regulated, a very weakly enforced space at the moment, people can essentially get away with many things. I think what we'll start to see in the future is technology like this and activities like this converging to make it much more difficult to carry out illegal activities or unregulated or unreported activities on the ocean. So for me, that's the exciting thing.

Satellites just acting like giant CCTV in town centres. So now these vast, expensive ocean, which previously felt a bit like no man's land, especially when you sail across them, there were protections for wreck, but they exist on possibly a name only, or sometimes. It'd been forced by someone going out and inspecting every so often, checking no one was around. Gone are the days of patrol ships. Exactly, right. But now you've got

this unimaginable tool. Yes, and so now it's actually cheaper and more effective to monitor remote sites because satellites can keep their eyes on and even the free data sources you can get from say the European Space Agency or from NASA which were designed for science can give you an understanding of what's happening in different

ocean spaces maybe once a week and so every week you might be able to get some free information about how this works. If not you have to buy satellite time you have to request that satellite looks at particular spot all the time, do you? Basically, yes. So there are different ways of doing it and different organisations have their own models. But if you want a very high resolution picture over a space, that's going to be very expensive

and there are only a few people who can do it. If you want sort of a broad brush over a broad area that won't allow you to really see the detail but gives you a clue as to what might be happening in different locations, that's more available. And then you have things like satellite radar, which can cover a really large area. And essentially what you're doing there is looking for the big metal objects. And they stand out really well, as do oil slicks. And that's a completely different model again.

I'm feeling a crowdfund coming on for a cultural heritage preservation designated satellite that only whizzes around and just checks out the cultural sites around the world at risk and although i would say i think that's probably not the most effective way forward okay i've cancelled that idea um simply because making and launching a satellite is not only very expensive, it has its own impact on the environment. And there are more than enough satellites already and more than enough people planning more than enough satellites. What we really need is to be able to get access to that information for these kinds of use cases at a more reasonable cost. Right. I think that's

the future. And there's no doubt that a there will be more coverage in satellites within a few years five to ten years we'll be able to see everything everywhere there'll be no doubt about that um and and and all the time in quotes um sort of maybe every hour you might see what's going on somewhere that is that is the ambition of some of the satellite operators now um and as you do, and more people start to use that data for things that they think of, then the price is going to come down. So I think that there is a natural trajectory

towards where we'll be able to get this data and be able to use it more readily for a wide variety of things. And then the other thing is access and capability to resources for processing, such as the Microsoft platform. This allows you to much more readily and cheaply get data, look at it, analyze it, put algorithms across it,

learn things that are happening in it for all sorts of different things. So I think we're going to see a ramping up of sort of tech for good entrepreneurs who are using these sorts of data sets for a wide variety of sort of SDGs and other societal purposes. Jessica, anyone listening to this podcast is a fanatical history and archaeology fan, so we don't have to convince anyone. But when you're listening to that, when you hear the costs involved, what is your argument to those who say, really, we spend all this money on preserving our maritime heritage?

Like, is this a good use of money in a world that has so many demands on us? Well, we all go to museums every day. There are thousands and thousands of people who love to go and see the wealth of our history, going back to the Bronze Age. And there's no difference between going to the British Museum and what's under the water, it's just that we can't see it and so it's unloved and underappreciated. But this country, as I said, it's got more wrecks than any other country on earth and that's primarily because we were trading from the Bronze Age.

We have so much history and we forget that just because we look out at scene we've we've just become a society of yachties and people who want to drink ice creams on the seashore but out there i feel seen but out there there's so much that we have to learn about um our past and there's so much wealth there that um enriches us as a nation. It's so true. I mean, you've obviously, better than I do, but every time I've dived off the coast, part of the problem is you're looking at shipwreck. There's usually another shipwreck on top of it, underneath it.

And I mean, that is often a problem. Absolutely. I think you could probably walk along the south coast from Brighton to Plymouth if you had long legs. On shipwrecks, yeah. There's the one in Salkham, the slave, the thing they might be a slave trader, a North African slave trader, which is,

well, you're rolling your eyes, but anyway, the one I was last there. But that's on top of a Bronze Age trading vessel as well. That's right, yeah. There are three of them. There are three, I think, aren't there? Yes. Yeah, on one spot.

Yeah. It's madness. Yeah. Or you look at the seabed, you can understand why they wrecked. Yeah, and also, I hope you're looking at that site. But one last thing, is this really a problem in British waters, though? Is illegal looting of our maritime heritage an issue?

as they see it. If they see something shiny on the seabed, well, I can have it. And there's, unless the wreck is protected, then there's nothing stopping them from raising said cannonball or ship's bell and putting it on their mantelpiece if they declare it in the correct manner. And so that's a bit of a shame because a large amount of our heritage sits in people's living rooms. because a large amount of our heritage sits in people's living rooms. And it is an issue because that heritage is then taken from the nation and kept for a very small amount of people and where it also will sit and crumble. It's usually large scale. I mean, one or two divers taking stuff from wrecks is neither here nor there.

I mean, there's been some high-profile cases in the last few years of cannons being raised and fraud being committed in order to keep said cannon. But it's usually the large-scale projects that we're interested in and which we monitor. There are wrecks protected by law under the Protection of Wrecks Act 1973. It's under 100, which is just a blip on the scale of what we've got off our coast and to some degree there's less regulation or or less um law about some of these wrecks because

it's perceived as difficult to enforce and so if you put a law in place that you know you can't enforce that's something that that is is very challenging for governments because they they will get challenged on it certainly would challenge now it's easier to enforce than most it's easy to force the most drug laws now with your technology because you're like don't worry guys i've got this certainly it's much harder to to look at things like the individual diver but for the the large scale where um the costs make make sense then absolutely, it's a doable thing. So it would be great to see more protections enacted and to consider the moral implications of people taking from what is essentially the common heritage. Well, that's, Nick, raises a very important point, which is to do with

war graves, because we have a legislation called the Protection of Military Remains Act, which is relatively complicated, but it protects warships and airplanes automatically around the world. But it's very difficult to enforce this, and if you're beyond 12 miles, nothing can really be enforced unless the captain or crew are British nationality. The other issue is because these are where thousands of men died and sacrificed their lives during the last two world wars.

But the Commonwealth War Graves Commission can only protect those in graves on land. So there is the difference between the moral authority and the legal authority. So you have a whole bunch of people who died, say, on the HMS Queen Mary from the Battle of Jutland, but they don't get a proper grave. Their grave is below the sea. So it's something that really needs addressing. But obviously the Commonwealth Grave Commission came in after the First World War when it was never even considered

that these wrecks would be found once they'd sunk sometimes hundreds of metres deep. But why should we forget them? Why shouldn't we look after those people who gave everything so that we could carry on? When those graves are on land, we are absolutely committed to our memory of those people. Why not when they were serving on ships?

But now your technology allows us to do that as easily, actually, as if they're on land. I'm not going to go with as easily, but certainly it makes a difference. You can definitely now protect things that were previously unprotectable without extreme cost. So it is definitely a much more capable environment that we live in. As ever, technology is galloping ahead of the political, the regulatory framework. Let's hope it catches up. Yes, well, I'm sure it will. It does in many other areas,

but you're right, it'll take time and it will take will. And yeah, getting this to the top of people's agenda is going to be the challenge, I think. That's really exciting, guys. Thank you so much for coming and talking about it on the programme. You're very welcome. thank you for having us i feel we have the history on

our shoulders all this tradition of ours our school history our songs this part of the history of our country all were gone and finished and liquidated one child one teacher book, and one pen can change the world. He tells us what is possible, not just in the pages of history books, but in our own lives as well. I have faith in you. This episode was brought to you in partnership with Microsoft. Thank you to them and for all the work they are doing

to help preserve our maritime environment.