 Good morning, everybody. Welcome to our panel on precision extinction. We have an extraordinary group of people here today. I'll introduce them one by one. I'll tell you a little bit about the panel is, and then we'll get cracking. So we have Marcos de Souza. He has perhaps the best job title there is, which is Secretary of Innovation for Brazil. We have Feng Zhang, who is a professor at MIT and the creator of CRISPR. One of the technologies we'll be talking about today. Many of you are surely familiar with it. We have Missy Cummings. She's the director of the Human and Autonomy Lab at Duke. Mark Benioff, who is the CEO of Salesforce, and my near relative, Peter Thompson, who is the UN Special Envoy for the Oceans. So what we're going to talk about here today, the World Economic Forum put out a report. And so one of the risks in the future is precision extinction, the possibility that the technologies of the fourth industrial revolution, so AI, drones, genetic engineering, could be used to not just protect the environment, but they could be used to wipe species off the planet. Came out in a recent report a couple of weeks ago. So what I want to do in this panel is to talk about those risks and then also talk about the ways these new technologies can be good for species, good for diversity, good for the environment, and then also how we drive technology in the right direction. How do we make sure we make the right choices about the ways that technology develops so we get the good outcomes and not the bad? Okay, so let's get cracking. I'm going to start with Mark. I'm going to ask you, tell me one technology that you're excited about on this topic and one that you are worried about. Okay, well, thank you for that. Well, I think one thing that I'm very, I guess I'll kind of bundle it together, but I think I'm going to want Missy to help me out. You know, I was recently hanging out in Hawaii on the beach and what I like to do is I like to go out and clean the beach a little bit. It's very meditative for me. And I was cleaning the beach and I had it all nice and pretty and fine and wonderful and got the plastics off the beach and the water bottles and I found an old gas tank in the shoreline and hauled it out. And then I was sitting there and very happy with myself and I'm looking at everything and I'm like, oh, I got it all perfect. This is fantastic. And then I just kind of put my hand down and I dug up and I noticed that all the microplastics were fully embedded in the beach. You know, that would be the right down, you know, as far as I could go. And I'm like, I don't have an extra lifetime to sit here and clean all this out. This is amazing. That as the ocean is basically bringing in all of this plastic and as we have more and more plastic in the ocean, it's getting embedded into our beaches. And then I was thinking, well, what I really want is I want a beach cleaning robot. So I want some of the great next gen vision technology, AI technology, robotic thing, and I can leave it there and it can go through all of those grains of sand and figure out what's a grain of sand and what's actually a piece of plastic, which I think it can probably do unless he's gonna help me out and clean that beach. And I want to put those little beach cleaning robots on every beach I can find and clean our beaches because some of our beaches are pretty disgusting, honestly. And I think that's one good thing. And I think on the reverse side is, you know, we're about to enter this new wave of deep sea mining. And if you haven't seen some of these robots, these are robots bigger in some cases than this room that are gonna basically go under the ocean. It's gonna be out of sight, out of mind. They're gonna be autonomous, AI based. They're gonna be looking not for microplastic but for valuable minerals and metals and things that we've exhausted, you know, on land. And they're gonna strip mine the ocean. And Peter's gonna make sure that doesn't happen. But, you know, that same technology, which is, you know, of course we're all here or we're all waiting for autonomous cars. We're all waiting for AI based cars. And all of that same technology can be used in these two different ways when it comes to the oceans. And so that's been on my mind. Those are great examples. Missy, do you wanna talk about one of your areas of expertise as drones? Sure, it's drones and more broadly, I do a lot of work across autonomous vehicles, including driverless cars. And it sounds great. We all would love the beach cleaning robot. But we actually have a robot that works in very similar environments. It's on Mars. Nicholas actually called the beach cleaning Roomba. Yeah, the Roomba. Yeah. So it turns out that, you know, while technology is often seen as a panacea, I think we overestimate in many cases its abilities. So sand, it turns out sand is just a destroyer of systems, mechanical systems in general, but they have very, we have a lot of difficulties with the NASA Mars curiosity rovers because they, it's just very difficult, especially when you're trying to operate remotely. But I would also tell you, while there are clear technical difficulties for a beach cleaning Roomba robot, my biggest concern would be somebody damaging it, humans. Humans damaging other robots. We've seen a lot of examples across the country where robots are left to do delivery or goodwill. And we're trying to see how humans interact with them. And often humans will destroy them, damage them. And so, and I've lived in, I lived in the Philippines where, you know, we would put technology out and then it would be scrapped for parts, right? And so these are in many of the cultures that we're speaking of, we have to understand the context. So it's not just species extinction, it's robot extinction. Well, you know, there's this, but then I will tell you on the flip side, that the thing about the deep sea mining with robots, you know, I think that it is, we should be concerned about this and talking about it, but my concerns aren't really with the malevolent use of these technologies, the intentional malevolent use, it's the accidental malevolent use. AI is definitely opening up Pandora's box. Most applications of AI, we still, particularly when it comes to autonomous vehicles, we really do not understand how the underlying algorithms work. And so I'm much more concerned about accidental damage that can be profound as opposed to malevolent damage. That's an excellent point and one I very much want to continue on, but let's continue going. So Feng Zhang, your technology, you help pioneers often talked about as one of the greatest hopes for species preservation and for humans and one of the greatest risks. So briefly describe what CRISPR is for everybody. Most people probably know, but let's do a brief recap and then something that excites you in this front and something that scares you. Sounds good. So CRISPR is a gene editing technology. It's a way, it's a molecular tool that allows us to go into the DNA of cells and start to make very precise changes to the DNA. So there are a lot of different applications for this technology. Imagine if we know what mutations cause cancer or which mutation cause metabolic disease, we can use CRISPR to go into cells in our body and get rid of that mutation and then be able to treat the disease. In agriculture, we can also use it to introduce traits that are beneficial, drought resistance, cold resistance, virus resistance. All these beneficial traits we can quickly and also precisely introduce into plans to significantly increase agricultural yield. So there are a lot of different applications and the one thing that really excites me is really the convergence of both our ability to read DNA and also to write DNA. The rapid progress that we have been making in the biotech field is DNA sequencing is becoming faster and cheaper. So we can read DNA at unprecedented pace. Genome editing is also becoming much, much more tractable. And so the combination of reading DNA and writing DNA makes it possible for us to transport traits from one organism to another organism. So a few years ago, there was a protein found from an Arctic species, fish, that provides anti-freezing purposes. And so this protein was transferred into strawberry to create strawberries that doesn't freeze. Yeah, inadvertently, that protein also turned the strawberry into blue rather than red. It's kind of a strange strawberry. But I think that the really exciting thing is that as we sequence more and more organisms, we can now find interesting properties that these organisms evolved to allow them to most optimally survive in their own environment and transfer some of those into other organisms so that we can improve the property. When we're facing with extinction of different species, imagine we can transfer the ability to survive into a different organism so that we can prevent the extinction of that organism. So if the tuna are going extinct, we can insert a gene that will help them survive whatever disease in the ocean is hurting them. The only risk is that it might turn them blue. That's right. Marcos, let's talk about Brazil. You've heard these three talk about some of the technologies they're looking for, some of the risks they see. Tell me how you think it might play out in Brazil for good and for ill. You have obviously rainforest, which the whole world cares deeply about. You also have long coastlines. Okay, well, in Brazil, what we are discussing, at least from the government perspective, is regulation. The characteristics of this fourth industrial revolution is the speed of the technology advancing. And as you know, the government regulation is always behind this technology advancing speed. So it's a challenge for us because the other previous revolution, they took longer, so we could prepare and adapt our regulations properly, but this one is going too fast. And for us as a government, not only in Brazil, but we are discussing here in the forum, for different governments, it's a new challenge because of the speed. Just to give you some examples in Brazil, we are talking about biotech knowledge. We have a national commission for biosafety that approves all the genetic-modified organisms. There's, of course, a lot of pressure for different parts of the society, but its members are from the scientific community. So this is a problem because sometimes, the time to analyze the potential impact for environment in this application of new technologies is too long. So it delays the adoption of new technologies in Brazil, and I believe in other places also, because they go deeper to know the impacts because if there is something wrong and the government approves that technology that goes wrong, there's a possibility it will be in the government. So the regulation about biotech knowledge in Brazil is critical because, you know, we are an agriculture power and it impacts directly our competitiveness. So this is an example. Another example about also regulation are drones. How do you regulate their space for drones? And in Brazil, we are seeing a lot of opportunities for drones in the agriculture to track all the crops, to know precisely where you use the agrotoxys and fertilizers and so on, but there's no regulation yet in Brazil. Some people are still discussing how to pilot drones, how to train people to pilot drones, but we are seeing that it's nonsense because all the drones will be autonomous. So it increases the complexity from the point of view of the regulation. How do you regulate autonomous drones? Because it's not one. When you go to Brazil, the crops are big, are hundreds of kilometers that they have to cover. So probably they will use a cloud of drones, autonomous to do that and to track everything, but there's no regulation yet. Let's talk about the rainforest second and the arms race. So presumably people who are illegal loggers who are chopping down the arms race can use satellite imagery to figure out where the best trees are. They can use drones to figure out the areas that haven't been logged. And presumably on the other side, you can track where those people are operating, right? You can track the drones. You can see where trees have been logged illegally. You can tag the trees so that you can see whether the crop is actually from where they say it's from. Who's winning that arms race right now, the loggers or the government? I believe it's the loggers because we are seeing a lot of development of this kind of technologies in Brazil. We see a lot of Brazilian startups partnering with Monsanto, Singenta and other companies. Making together the seed development but integrated with drones, with sensors, with IoT technology in order to track that. So they are not waiting for regulation, of course. They are going in front of that. But at some point when it increases a lot and it scale up, probably we'll have to do some smart regulation. And so Peter, I wanna ask you a similar question because with the oceans, there's gotta be a similar arms race, right? There's gotta be people who are, you know, the fishermen hunting the most valuable fish are using satellite imagery. They're using technology to bring the fish out past the 200 mile border so they can be hunted without jurisdiction. But of course we're also using technology to track tuna and to make sure that places that buy tuna are buying tuna that was caught legally. So tell me how as somebody who's got more experience regulating the oceans than anybody else, tell me how you can help set the regulations so that the side of preservation wins the arms race, not the side of extraction. I think there's a lot of good news to give in that regard and I would look at the glass half full rather than half empty in spite of the fact that the ocean is being put under most awful pressures by humanity, whether it's- Half full but a little acidic maybe? Well, just that everything that has been said by the panelists so far to me are half full responses because, you know, for example, robotics, robots might not look like sand but they're like water and there are 10 rivers in the world which produce according to some stats 90% of the plastic pollution in the ocean. I mean, surely there must be robots that can go to those 10 rivers and clean up and deal with 90% of our plastic pollution right away there. On CRISPR, as you were talking, I was thinking last week I was in Fiji launching the International Year of the Reef with Eric Solheim on a third biggest reef in the Southern Hemisphere, the Great Sea Reef of Fiji and looking at that coral and just the tragic thought that, you know, if you believe people like Sir David Attenborough, there will be no coral reefs by the end of this century. But what we're finding from people like Wood's whole research is that there are resilient types of coral. So if there are resilient types of coral, surely CRISPR could develop those resilient or change all coral to be resilient and we could save coral reefs. So, you know, I see that in a positive way. If there's a downside to that, somebody tell me, but I would see that as a very positive thing. As far as governance is concerned, you know, the United Nations is sluggish as you all know, but in December, you know, a month ago, there was two really big movements. One was the approval for what's called BB&J and that's Marine Biodiversity Beyond National Jurisdictions. That is basically to get law in place to end this piracy zone that covers the majority of this planet, which is the high seas. The conference will start in September, it'll last for a couple of years. Law will result. When you start a UN conference, you end up with law. The question is, how good is the law? But there will be law resulting and that will be fitted into UNCLOS. So good move there, the piracy era for the high seas is gonna come to an end. Secondly, we know more about the face of the moon and Mars than we do about the bottom of the sea. And the UN approved the international decade for ocean science for sustainable development. That was approved in December. It'll be run out of the IOC, the Intergovernmental Oceanographic Commission in Paris and it's a decade where for the first time, humanity is really gonna focus on ocean science and that runs from 2021 to 2030. So there are good things happening at the international level. Just finally though, since seabed mining was mentioned and yeah, it is a horror story to a lot of people. I was president of the International Seabed Authority for two years. I can tell you that seabed mining is definitely coming but I can also tell you that it's not allowed at present. Why? We don't have regulations at the International Seabed Authority to grant licenses. But again, that's law which is under preparation now and we'll be ready soon. But it's a law which has to be approved by all the signatories of UNCLOS. So it's not cowboy territory. This is probably the best-governed part of the planet is the seabed because it's governed by UNCLOS and the International Seabed Authority. But what I've been ramming home to those guys is we must lift the bar very high on the precautionary principle when it comes to that regulation if somebody wants to go seabed mining. They gotta be able to prove to us they're not going to kill the goose that lays the golden egg. Mark, do you agree? Are you confident that regulations will be able to prevent seabed mining and are you optimistic that the government can prevent some of the worst technological things that happen in the ocean? Well, first of all, Peter's doing an amazing job and we're less so lucky to have him as our UN Ocean Commissioner and that the world is starting to work on the oceans is so important. I think that it's been ignored for far too long and it's great that we are moving forward and like what he's mentioning about governance of the high seas, I mean, it's the wild west still and that he's leading the charge to rein it in. This is great news for the oceans. And when you look at the major ocean issues, things like acidification, which is seas are getting warmer, global warming is creating a warmer ocean it's gonna change the biodiversity of the ocean. It's happening, we were talking also that there was a panel here at the conference on led by some very pioneering Chinese scientists on bacteria in the ocean and kind of a revelation that the ocean if you will kind of has a biome like the human being has a biome and that this bacteria that they were talking about that they're sampling is at this 25,000 foot level and that how it can hold or control methane release. That was an amazing session here with David Agus. When we look at acidification, when we look at overfishing you mentioned overfishing certainly low hanging satellites which are emerging and we're about to have thousands and thousands and thousands of real time satellite imaging that's gonna dramatically reduce overfishing because we're gonna know where every fishing vessel is in real time across the world. We're not gonna wait for this idea of beacons or whatever in the ships, we're gonna have that information right off the satellites and plastics, soon we're gonna have more plastic in the ocean than fish, this has gotta be dealt with and I hope Missy's gonna come up with some robots for us but we also saw Boyan Slat is here at the conference who runs the ocean cleanup, he has his big experiment where he's gonna try to create a passive system for cleaning the ocean but this is a big deal like Peter said that 90% of this ocean garbage is coming from these 10 rivers and I mean everyone, if you've been to the beach anywhere in the world, you can see it. Now I was just in Brazil also and just kinda coming back to all of our role and one thing I think that's very important about the Fourth Industrial Revolution and then we talk about this a lot in our center in San Francisco is cultures and how cultures and consciousness also we're gonna have to change in regards to a lot of this technology, the awareness. It's one of the really important things about a multi-stakeholder dialogue like we're having on stage here or that's happening at the WEF but when I was just in Brazil and I was incredible country, maybe the most beautiful country I've ever been to and I was out on the beach at night enjoying the sun coming down and I noticed that some people are bringing their garbage to the beach because they're waiting for the ocean to come and take the garbage out and I'm sure that's not true of everybody in Brazil but I was just noticing like wow, that's not happening where I live, it's a different culture, different consciousness, we need to somehow get everybody on board if we're gonna save for example, save our oceans. I wanna, you had about 19 extremely interesting things that I wanna focus on one in the middle which you said the ocean has a microbiome and that we can engineer the microbiome perhaps to change methane absorption or for other reasons. That sounds great, that sounds terrifying. So someone else wanna talk about the risks and rewards about messing with the microbiome of the ocean? Sure, I'll jump in. I do think we need to be careful about engineering microorganisms and releasing them out into the environment. Biology is really complicated. Usually within a cell there are thousands if not tens of thousands of different genes that play very complicated interactions with each other. When you change one gene, predicting that it will help reduce methane or try to improve some property that will be beneficial for us, it's hard to predict how that one protein would interact with all the other things in the cell and change something else about it. I'll give you one example from the human therapeutic side just to give you a sense of the complexity. So about single digit percent of the human population carry a genetic mutation in a gene called CCR5 and those individuals are immune to HIV infection. So you may think that this is a pretty good mutation. Why don't we just vaccinate everybody by removing CCR5 from everybody? Turns out those individuals who are immune to HIV they have an increased risk for West Nile virus. And so even though we don't have a West Nile epidemic right now, if we prophylatically install that mutation into everybody, then if the virus came then we'll be really in serious trouble. So when we're engineering organisms in the microbiome in the ocean, especially something that really affects the global environment, I think we have to be very careful and proceed with a lot of caution. Are there other? I think to add to that we were talking before but recently we were having some Zika issues in the United States and there was definitely a very high level discussion in the government should we use CRISPR and gene drives to wipe out the mosquitoes so that we got rid of the carriers. And the world is a complex system, human beings are complex systems. And one little change, you're gonna have unintended consequences and I think that's a very important. Okay, but let's go down on this a little bit. So we have to be cautious, totally agreed, but can we be more precise about that? So let's say that there's a technology that we have a pretty good chance we'll reduce global warming by sucking down carbon dioxide or reducing methane releases. And we have like a real crisis with global warming. So at what stage do we say, you know what? Let's just do it, let's try it. Let's re-engineer the ocean. I think one of the things that, especially for biological systems, you'll be important to engineer containment mechanisms. So rather than doing something that's irreversible, can we engineer a circuit in these biological organisms so that once we release it into the environment and just in case something goes wrong, we can switch it off. We can recall this strategy back. Miss, you think that's a good idea? Yeah, I was just thinking to myself, we already do that in engineering, that's why when you fly commercial jet, it's so safe, it's because they're triple redundant, right? So we have two backup systems in technology. Again, though, I would actually say the issues that you see with CRISPR in terms of it sounds good, but there are often these unforeseen problems that emerge. It's the same with AI. We see, yes, there are some applications that can be good, but it turns out for driverless cars, for example, right now, one of the things that was just recently discovered in the last six months is that using very easy passive hacking techniques, i.e. just putting a few stickers on a stop sign can trick computer vision from seeing a stop sign and makes it see a speed limit of 45 miles an hour sign, right? And so we had no idea that these things were possible until just the last six months. And so as a researcher, what I worry about is, well, if we're still finding out these emergent properties of these technologies, CRISPR, AI, yet there are many companies and agencies that wanna take these technologies and start deploying them in the real world, but it's still so nascent that we're not really sure what we're doing. So I do think that it needs to be more of a collaborative arrangement between academia and governments and companies to understand what's really mature and what is still very experimental. Well, Peter and Marcos, do you wanna weigh in on that? What is the role of government in regulating these technologies and setting policies? Because often government trails behind the technologies and understanding. I was just thinking, as the panelists were talking, that geoengineering came up there. The ethics of geoengineering, this is not something that's discussed. In fact, geoengineering is not discussed much, but as somebody rightly pointed out, it needs to be developed as a sort of backup plan, but I think it's the ethics of it. When I was president of the General Assembly, I tried to stimulate a day of discussion about ethics of innovation. A lot of people came over from Silicon Valley, and I tried to put government and Silicon Valley together for a day to just start that ethical conversation, because it is missing at the moment. I mean, you have it in the United States for something like CRISPR, I think has to go to a Senate ethics committee where they discuss it and so on. But for globally, I think this is definitely lacking at the moment and is needed a global discussion on the ethics of artificial intelligence, genetic manipulation and so on. And I'm sure the scientists would appreciate it as much as the governments would. Yeah, actually we will answer with a question to Mark Benioff. One of the challenges for us in the government when professor is developing a technology are the impacts. How do we simulate different scenarios more accurately, more precisely, what are the impacts that we'll generate? This is a long process, involves a lot of scientific data and so on. How could we use IT technology and computing power and artificial intelligence and whatever to create and to be more precise on the impacts that could generate? Because this discussion take longer, how could we use technology to use scientific data, evidence-based decisions faster and more precisely? Because it's important to try to separate what is ideology because there's a lot on this field from evidence-based decisions. That's all my question. That's a good question. Is that to accelerate? Yeah, I think that we have the opportunity today and I'm gonna also link back to Missy, but there's more data than ever and there's more data coming from all these sensors from not just all aspects of the environment, but we're talking about DNA sequencing, that massive amounts of new types of data available. And that through that, we have the ability to have analytics and insights that we've really never had before. We're in a data revolution. I think coupling it with artificial intelligence, that's where it can get exciting because human beings are not gonna be able to sort through all of this data. It's just, and all of a sudden we're looking for AI maybe to help us or to guide us or to augment human intelligence to have that insight based on all this incredible new data that we have. I think that's very much your answer and I think when you look at all the data, you mentioned all the drones you have and the sensors in the environment and all that. When you bring all that in, I think you're gonna be able to have some tremendous insight. I'll give you a very kind of basic business example. At Salesforce, my company, we're using AI in a way that we've never used it before and we haven't released it yet to our customers because we're still not sure about it, kind of to Missy's point. And how we're using it is we have large databases on how we run our company and how I run my company as a manual process is every Monday I have a staff meeting, like I'm sure a lot of CEOs do and I have my top 30 or 40 executives around a table from, it's a virtual table. There's people from all over the world as well as physically with me and we figure out how we're doing as a company. We're looking at all of this analysis but now I have a new person at the table and it's kind of an empty chair and we have a technology called Einstein and I as a CEO ask Einstein, okay, I heard what everybody said but Einstein, what do you actually think? And now that I've been using this technology for well over a year, in each and every time I ask it, it always has an insight about an executive or a territory or a product that I would never have seen. There's too much data for me to understand what's going on and also I have executives like I'm sure a lot of people do who are holding back information or gaining what I call gaining information from me. Things they don't want me to know because maybe things aren't going that well. Einstein has no bias. So Einstein says, well, like I had a situation recently with a European executive and I said, well, I don't think this European executive is gonna make their number, I'm so sorry. And then the European executive just got so upset and they're like, no, I am and here's why. And then I said, no, I'm sorry, you got this problem, this problem, this person doesn't have this in place, you didn't do this over here. And it's looking at years of data. I think it's a good example actually because certainly like when we have physicians and medical professionals who you walk into your doctor's office today, your doctor does not have that full understanding of all of your data, right? And AI is gonna be first and foremost, I think in the example I just gave a partner. In the medical example, and this goes directly to your question, today when we ship a CT scanner, it's a dumb device. It's not really intelligent. You jump in it, it does a scan, maybe you do a scan of your heart or your body or your brain or whatever you're looking at. And then somebody called a radiologist comes and looks and says, oh, what about this? What about that? What about this? But in the next generation of the CT scanners, it's not gonna be a dumb device. It's gonna be a smart device and it's gonna be a diagnostic device. That's the leap. And that's actually what you're looking for. That is what the CT scanner is doing is it's generating all of this data, like what's happening in your country with all of your innovation. And then you want that AI, basically, in the scanner to say, oh yeah, that has a hepatic liver. And we better go look at that. Instead of waiting for somebody who knows what that looks like to come in and say, oh yeah, this is my diagnosis. And I think that is our next leap in the short term in technology. Things that are previously devices, technology, robots, cars even. We don't have to make the full leap to autonomous vehicles to have smart cars. So let's not, you know, we can, and we're gonna be, we can be augmented through artificial intelligence as human experts, as doctors, as CEOs, as government officials, and also global governance. You know, I think that we have to really come back to you because global governance, I think it's heightened in this environment. It's far more important than it was a decade ago. And so there's more of a burden on you. Would you agree with that? Well, I do, because when it comes to the ocean, there's only one ocean, and it's no good fixing everything in Brazil and leaving everything else unfixed because as we know, things like plastic migrate across the ocean. Fisher don't recognize national boundaries, et cetera. So, no, I mean, the more I hear the panel talk, the more I'm confirmed in what I said originally that I'm a glass half full on this. I think innovation and innovative technology is gonna be a huge part of the global solution to the reversing the cycle of decline that the ocean's currently caught in. Whether it is on the fisheries sector, I mean, there are obvious applications of innovative technology there. Cleaning up pollution, plastic pollution, we discussed that. The decline of coral reefs, you know, Chris could have a role there and so on. So, rather than seeing rogue technology as something that's gonna get out of hand and destroy the life in the ocean, we're doing a very good job of that without rogue technology. What I think we can do is use technology to help us in reversing that cycle of decline. I'm very confident about that. Missy, do you agree? Are you more optimistic now or at the end of the panel? I am. I mean, I'm a futurist and a technologist. Clearly, I'm gonna be positive about the future. I think it's worth looking back though. 10 years ago, this community, even here at the World Economic Forum, but more globally, were very anti-drone. So, I've been actually doing drone research since 2001. That's a long time. So, most of you didn't even know drones were a thing in academia in 2001. And so, and I saw a huge backlash against drones. Really, up until about 2013 was the big pivot point. And it's rare that you can look at a point in time and say technology has culturally changed. And that's when Jeff Bezos made his announcement that Amazon would start delivery with drones. And it was that moment that caught the public's attention to understand, wait, wait, wait, this technology that's gonna kill us all is now going to be used in a different way. And it's interesting, over the years, I do a lot of interviews and they were all very pointed in a very negative way about drones. And now, that conversation is very rare for me. I only rarely have to defend the use of drones in military settings. Most of the time, we're talking about commercial applications. And in terms of conservation, I think one of the interesting statistics is four non-military applications of drones, one third of all drones are being used in some sort of conservation effort. So I work with companies who do drones for tracking elephants, for example. And this is a good example we were discussing earlier that drones can be very good at tracking wildlife. The higher they are, the better you have. One of the flip sides to this is, turns out elephants do not like drones. They run, we've had elephants throw mud at drones. And it turns out, this goes to that unintended consequence, it turns out the research we've been doing at Duke is, drones are on the same frequency as bees. And elephants hate bees. That's really their only enemy in the wild. And so, you know, but should we continue to use drones to track elephants? Yes, we should. It's still a very valid technology, but now we need to understand we need to do it higher. We need different kinds of engines. Maybe we need to change something we call our concepts of operations to make sure that we're not doing more damage by using them. But that doesn't mean we should eradicate them. Or can you get Jeff Bezos to talk to the elephants and maybe they'll flip, too? Yeah, but you know, it's interesting this idea of the technology, I call it the drone space race. The poachers now have drones, which they're ordering off of Amazon. But then, of course, so do all the conservationists. And so I think you will never win this battle. There's never a, which is better, which is worse. The technology is just a tool. So in the end, what we need to do with the technology is, and we were just discussing this, it needs to augment humans. We need to learn how to bring it in as a tool and not a panacea. I'm gonna move that. To add to that, I think that we touched on this briefly, but I'd love to get your thought on that, which is that these next generation of low-flying satellites. So we're, there's companies now that are launching tens of thousands of low-flying satellites, and you're gonna have that kind of imaging that you're talking about or that you're looking for available to you real-time imaging of so many things that today, or maybe limited by that drone, you'll have covered on a big swath of geography, not just for fishing and fishing surveillance as a great example is we have to surveil the ocean to see if somebody is fishing in a marine protected area. For example, in your case, you wanna make sure that there aren't people heading to those elephants. I think with low-flying satellites coupled with artificial intelligence, soon you'll have that early warning system that you just do not have today because you'll know before they get there that something is about to happen, that there's gonna be some much greater awareness created by that real-time imaging. I think that's correct, yeah, I think it's correct, and I think what people forget are that drones, there's nothing special about a drone, it's a flying camera, and the reason that they've become so popular is that access to satellite imaging right now is prohibitively expensive for most companies, and so I don't think it means that we'll get away from drones, for their ability to do local response in real-time, it will be unmatched, but certainly for these long-term monitoring situations as satellite technology improves that will, and the cost comes down, that will become more available, but one of the things in the ocean discussion, the ocean debris, just behind ocean debris is space debris junk problem, and so actually we're trying to come up with space robots now to try to clean up the debris field in space, which is substantial, so not nearly as imminent, but also if we're gonna go to these massively numbers of low-flying satellites in orbit, we actually just don't have the room right now. So we need sand cleaning robots, we need ocean cleaning robots, we need space cleaning robots. We have about 15 minutes left, I'd like to open it up to the audience, we have five marvelous panelists here, raise your hand and ask some questions. Yes, right there in red. Yeah, I think it, actually. So the live stream can hear, the question was more details about Einstein, and I should note for all of the questioners, if you could take the microphone, because that will make it better for our web audience as well. Okay, Mark. Yeah, I'll try to make it more generic to artificial intelligence in general, but of course we have very large datasets in our company, and that very large datasets for an average person, CEO, I'm just, you know, how do I make the best decisions? And the way I'm gonna make the best decisions is by being able to take all that data and do that. I think in every single example we went through here, there's more and more data. That's the big challenge. It was one of the big challenges of the fourth industrial revolution is more data than ever. When you couple the more data then with AI, then that becomes an opportunity to augment human capability. So that as a CEO, I can ask a question of Einstein, my virtual management team member, and say, well, how is the company doing? Are we gonna make our quarter? How is this product? What geography should I travel to to have the biggest impact for the company? For a doctor, it's exactly the same situation for the person who runs the oceans. It's exactly the same situation. For him, one of the greatest challenges is he's done an incredible job and all of his associates over the last decade in creating these marine protected areas. It's one of the great accomplishments, I think. But are we surveilling them correctly and really making sure, well, there's a lot of data out there, we need more data, then we'll couple it with AI, then the AI will call him, his Einstein, whatever it'll be called, we'll call him and say, hey, there's a boat heading into this area, and you gotta send your patrol out to fix it. And I think that's the next step for a lot of these industries and a lot of these challenges. Especially with a lot of the MPAs are very remote, like the Pitca and the Henderson. Yeah, his point is that some of the areas of the ocean that we're trying to protect are the most remote areas with the most pristine areas and the specialized species and highest levels of biodiversity that the UN is just saying, well, let's make sure we actually keep this perfect. Well, if all of a sudden there's some big trawler heading into that area, we need to make sure he gets a text at three in the morning and that they send there, notify that local government, whoever it is, to send that patrol out and stop that boat. Today, we can't really do that in real time with that kind of efficacy, the same efficacy that I'm mentioning when I'm running my company or the same efficacy that a doctor can in terms of diagnostic or capability. But that is the short-term leap for technology. That is within the next few years, right? Would you agree? Oh yeah, absolutely. Yeah, this is where we are today. We are right there. And I think for a lot of, like I said, we have it now internally. I think one of the reasons that I had good performance as a CEO is I have this kind of technology and I want to make that available to all my customers, but I am cautious because it has not been all perfect. There have been issues. We're in a world and that's another example where it's like, okay, I don't want to turn this over and then get a call from a CEO that he or she made a bad decision about that because we didn't have this exactly right yet. Well, I happen to know from the green room that Peter was up at three in the morning so he'll get the text from Einstein. We had a question here in the front row. Thank you, Espen Martaida from Norway. It's a fascinating discussion. I think the clear message we're getting is that this technology is coming or these technologies are coming. They're not gonna go away. We have to relate to them and there's a need for some kind of normative conversation on how to deal with it. So my challenge to anyone in the panel is what exactly is the place and the content of this? What do we want to regulate or we want to put the normal normative framework on it? But what should it say? I mean, it's clearly it's not gonna say stop technology from developing, stop good uses and is it purely in Peter Thompson's in the public domain? Is it the UN or is it in some kind of a more interactive framework between the technology driving industries and the public regulator? Excellent question. Key question for Davos. Who wants this one? Well, I want to just take it from Davos. As a trustee at Davos, I would like to just say one thing about that, which is that this is the purpose of this conversation while we're here. But this idea that Klaus Schwab, coming out of UC Berkeley in the late 60s, we don't know everything he was doing at that point, but one thing he did do is wrote an incredible paper on stakeholder theory. This idea that through a multi-stakeholder dialogue, like we're having up here, we can elevate consciousness and solve and or work towards some progress. And I think it really evolves. The conversation we're having here, we've now built a center in San Francisco in the Presidio, which is what the World Economic Forum to have these dialogues and we're having these dialogues with the very technologists that are creating it and like to invite everybody in this room to join us at that center for continuing these discussions or bring your management teams or your colleagues to that center for that discussion. But I think this is the very, this is why and one of the reasons I love Davos have been coming here since 2002 is because I think this is probably one of the only places we can have this dialogue like that. Exactly like this. Anybody else want to take a crack at this? Yeah, I agree with everything Mark just said there, but specifically to your question, I think the regulations of this subject area we've been talking about lie principally within national boundaries. And by the way, Seabed mining of course, the countries have their own EEZs and can do what they like with them. But it's principally the regulations within countries. But where the global discussion, where the Davos discussion, the UN discussion is really necessary is the ethical side of things. Because of course if one country gets a certain technology or is aware of a certain technology, let's take CRISPR for example. And my country might say, no, no, that's playing God, we don't want that in our country. Well, there'll be people in our country who will also point out, but don't you realize that that's just putting us at a competitive disadvantage because if it's happening in that country, that country and that country, it's happening for the world. So that's the limitation of just taking a national view on things. So that's why the global discussion, particularly on the ethics side is really important I think, yeah. Yeah, and the trick question is how to balance innovation and development and regulation. And that's the third event that I participated in the last two days that's discussing exactly the same question. And but there are some solutions that were discussing this issue. One is the role of the Fourth Industrial Revolution Center in San Francisco that develops projects to accelerate regulations about these new issues. So Rwanda for instance is developing a drone airspace regulation based on this experience for the Fourth Industrial Revolution. Another option that was discussed was about having a regulatory sandbox where the government could try for a short period of time and try this regulation and change it faster. So this is one of the options because it was completely agreement among governments representatives here in Davos that sometimes we spend four or five years to regulate something. And when we regulate the put in place everything has changed. So we have to begin everything from the scratch. Yeah. All right, we have a question in the second row please. Thank you. My name is Wouter from Holland. A lot of the amazing technologies that have been discussed could qualify as geoengineering almost. And especially with regard to combating climate change I think geoengineering is becoming a larger part of that debate. And I'm very curious about the panelists view on concrete technologies that could help combat climate change and maybe through geoengineering. I'll just say that that's a great question because in my, we all live in little filter bubbles. Perhaps the most divisive issue is geoengineering because so many people feel like we've got to do it. We have to do it now. It's the best hope. We have to shoot sulfates into the atmosphere. We have to pull down carbon dioxide. And then other half the people I work with and now are like, well, I don't know. You start even talking about that. Then we'll stop reducing, reusing and recycling. It's an awesome issue to discuss. So experts, what should we do? So I think on the biological side, I think it's quite promising because when we talk about engineering biological organisms and then releasing that into the environment because we're starting to learn how to program biology we can begin to build these controllable circuits with redundancy so that when you do release this geoengineering solution that's biologically inspired and based, it's possible to have the ability to reverse it. That's right. I wanted to ask you this earlier, but we've been talking about AI and we've been talking about AI at levels that the humans don't understand. So this is the wonderful thing about where AI is going where Einstein will make predictions and you don't even know how Einstein made the predictions. So as we reach that level of sophistication does it make it harder to build in a kill switch? Like if we're not sure how the system's operating how do we know we can turn the system off? It's a good question. I think AI will help us understand more because as we collect more data both just the sequences of biological material but also the interaction of how biological programs work. AI will help us understand and be able to predict if you perturb one element of this complex program how is that going to affect overall execution in that organism? So I see that as a positive thing. Well to add to that, you don't have to go any farther than look at the elections and what happened with social media and what's happened in the last 90 days and social media CEOs talking about the unintended consequences of social media technology. We don't have to make the jump all the way to CRISPR and AI and robotics. You can just look at the current news stories and front page stories of CEOs who make this statement. I didn't realize that our technology could be used like that. We just didn't understand the technology that we built. We didn't understand that that could happen. Am I right? Isn't that the story? That is the story of the last two months in technology the story of the last two years in technology the record is the big story. That's an example right? So that's we're not going to the most sophisticated we don't have to go all the way to gene drives. Right, we don't understand Twitter. It's true. I mean it's a great point spot on. So I think that that's an illumination. I think the other illumination has to be have to come back to here where, you know, when we look at artificial intelligence the ability to have this dramatic impact and we talk about it making us healthier, wealthier, smarter. I think you have to start to ask is this about haves and have nots when it comes to AI, especially in regards to global governance. You know, when I walk into the UN we talk about all the human rights. Right, isn't there so like a big wall that talks about these are the rights of human? Is that right? Absolutely. Yeah, and so I wanted to ask you based on what you're hearing is AI a basic human right? Should it be on that wall? Should all human beings have access to this technology since this is going to be so important going forward? You know, I think Mark you're touching on something that is absolutely at the core of everything I hear coming out of Silicon Valley. Made a trip over there as I mentioned. And I came away from there seeing a new vision of the world. Yeah, you know, I drove around in the driverless cars and I saw the AI and what it could do and all the rest. A week later I was doing a tour of Africa. You know, I was in Central African Republic and Addis and things and I was in the congestion of camel carts and trucks, huge trucks and all those different forms going through pothold roads and everything. I would be so embarrassed to stand up in the town square and tell them what I'd just heard in Silicon Valley because it just patently was just for a rich set of people in the world, a small percentage and the majority of human beings will never get anywhere near that stuff anytime soon. So it's not something you could even discuss over there. So I think we have to be aware that we cannot be shaping a future where that would be the dystopia for me. That half of us or a quarter of us live in this amazing world where we live forever and the rest of us are peasants sitting in our countries like Fiji and Africa. So does equality then become a major discussion point in regards to all of these technologies? That's why I keep coming back to this point about ethics. I think that it should be in the global commons that we have that ethical discussion of where is AI taking us all and employment, all those arguments are common problems for us, not just for rich countries. So we have time for one last question. So here in the second row. I have a question for Douglas Macaulay, University of California, Santa Barbara. Question for the academics and the representatives from governance. So companies like Mark's are on the cutting edge of developing new technologies to streamline efficiency and problem-solving business. Are these technologies that are being developed in industry getting to researchers, conservationists, getting to governance, getting to managers of these systems fast enough? And are they there for application and use? I'd like to just answer a slight derivative of that because I think it will help you understand the broader context. We are in a global AI crisis right now for talent. We cannot, universities cannot put out enough people who understand AI. AI is many things. It's actually a very broad umbrella. AI is very good for a lot of what we heard here is post-hoc analytical reasoning. But that's very different from AI that controls safety critical systems like cars and planes. And we don't have enough across the spectrum. So you say, are these good lessons learned getting to the companies? The answer is probably not because companies are running so fast on the AI hamster wheel that they can't keep up. We can't keep up enough in producing the people. And this compressed timeline for education means that we, me, other professors here, we're producing people that probably are not getting the comprehensive education and AI that they need because we are working to get them out so fast. Education programs are very archaic. We want to stick with the way that we went through school. We still train people in the university setting globally like we did 30 years ago. And that's just simply not a good model anymore. So there's a lot to be done, I think, in the education sphere. Yeah, I think that's a really good point. And I think it's really not just AI. It's really more broadly, for highly skilled trained people coming out of academia, we are not producing enough of them quickly enough. From the government perspective, we are discussing exactly this yesterday. First of all, governments are conservative and no risk taken. Business is the opposite. The challenge is, how do we use technology that, as you mentioned, developed for business, for the government use? The first step is the lack of knowledge. The most part of the government is in developing countries, but also I have colleagues from developed countries that have no idea about anything that we are talking about, about the potential of this technology. So the first point is the lack of knowledge. The second one, OK, if they understand what are the possibilities, the second question is if it's right or wrong, but it's a risk take. So it's a responsibility from the government if they adopt some kind of technology in our early stage. What are the responsibilities if it goes wrong? For the public opinion and for the auditing bodies that make auditing for government. So we have to develop some kind of pilot projects that's one of the discussion that we had where companies could join the government, sometimes for free, just to develop and showcase the opportunities. We have some experience like that in Brazil where companies joined a specific NGO in Brazil to test new technologies or consulting problems and they donate to the government the project, in a pilot phase. After that, you measure, you approve the point and so you make comfort, make comfortable for the government to have a public procurement dedicated to that. So these are the main challenges in face that not only Brazilian government, other governments also face, the lack of knowledge. How do we implement this technology with a risk environment? And third, what are the responsibilities and how to scale up this project if it works? So I think we should develop a more close cooperation between government and business in order to solve this problem that you mentioned. All right, this panel has a kill switch. I've got to exercise it. Thank you all for coming. I would recommend that next year, the same five panelists be invited back but we do it with six and we add Einstein. So thank you all for coming and thank you for a great panel. Thanks.