 Our next presentation is from Beijing, China with Chong Dai Chen of the China Biodiversity Conservation and Green Development Foundation, who is going to explain how the River Eye app integrates AI with biodiversity conservation through bottom-up environmental governance. Welcome again, Chong. Hi there, ladies and gentlemen, welcome to our transition. My name is Zhang Dai Chen, I'm from the China Biodiversity Conservation and Green Development Foundation. Today I will be introducing River Eye, an app that incorporates AI with biodiversity conservation via bottom-up environmental governance. My presentation will be in four parts, its background, the app's introduction, its inner working mechanism, and finally the data analysis. Let's go one by one. The key points I'm trying to make here is that only by using technology can we effectively strengthen the public participation, and only by using big data can we strengthen the aquatic ecological protection. So, a little bit for the background of the app. Currently, China has 2,071 marine protected areas, but only 4.1 of the ocean was protected, and that's far from what the CBD Co-15 has designated. And also, China has just passed a law on the 10-year-old fishing ban on the Yangtze River, which is a great news for all the environmental protectors in China. And there's plenty other measures like banning extensive fishing measures, banning explosive fishing, etc., etc. So I see to say that the situation is going on pretty well in China. Even though there's plenty legislation just came out about water protection, there's still plenty of problems with the law enforcement and problems like this. We have lack of train staff, we don't have that much river chips, and the areas in China are too large for an overall supervision. We have scarce equipment for radaring and searching the illegal industrial features. And we have lack of law enforcement because it's water areas, they're planning to catch. But that's the problem before this app came out. So, the core mechanism of the River Eye app is basically about enhancement and big data. As for enhancement, it's mainly we serve as a platform of connection between government and the people, so that the government can work more efficiently, and people can report more efficiently as well. And as for big data, we have this mastermind-like platform that can do big data analysis to influence the government decision-making afterwards, and we can provide data for the future policy. As for the data collection and the mechanism, it's quite complicated, but mainly came down to two parts. Once a user has spotted illegal ads, whether it's sand mining, polluting, or electric poisoning, or explosive fishing, they will all be first filed by the AI to determine whether it's a false report or not. And then, after second-handed human classification, it will eventually came down to the governmental part, and we will tell them whether you should act on it and how you should act on it. This has cut down a lot of human resources, when it comes down to collecting reports and coming down to the sites, because River Eye already does the observe and report part for them. As for the data coverage, we have about 50,000 users in total, which I think is a pretty big number. They came over around 27 promises. Some of the promises don't have large water areas, so they're not included. And just for the past three years, we have submitted about 9,000 valuable clues. And we have direct cooperation with 105th local governmental units, for example, Fishery Administration or Public Security Bureau, et cetera. So let's take a look at the control page about how we do the real-time transmission with authorities. First, once a user sends a report that was identified as valuable by the AI system, this will first show the river chip how it looks like in that place, whether you should come to it with the right equipment or you should bring more people. It all can be done via this terminal. And then with the help of Baidu and some other maps, we provide them the most optimized road finding. And finally, after they have sent in the reports after they've done the investigation, we will let them have the feedback directly to the local government, which is far more convenient than the paperwork report sessions. Finally, I will give you another shot about what the department is seeing. The left picture is what the Fishery Bureau of the Ministry of Agriculture and Rural Affairs is seeing. As you can see, the purple indicates electric fishing which are popping out all over the country. And the right one is that it's what the Public Security Bureau in Chongqing is seeing. Some of the green one suggests that the case has been taken part of. And the purple one suggests that it still needs to be investigated. So finally, let's come to the data analysis part. Here are just a few of the data that we have taken for last year. We can see the illegal fishing cases coming up and down, whether the electrifying is taking a day or a night, and some of the cases that's all over the country we have taken in September. And finally, the classification of crime, etc., etc. These are all valuable information for the government to absorb and use in their future decision making. Finally, I'd like to give a conclusion about the solution to the previous problem before the app. The lack of staff was solved by low interest volunteering. As long as we have this app and you take a pic, the platform may take care of the rest. And as for the fake reports, we have AI system for reports filtering. And as for scarce equipment, our phone isn't up, we don't have to bring cameras, we don't have to bring drones. The warranty here is OK with just a phone. And finally, since we have got a connection with local governments, we actually have coordinated law enforcement without all the paperwork, without all the second-handed information. And that will be all. Thank you for the listening. With public participation, we can all become AI for the river. If you are interested in the topic, you can follow us on our social media. And that will be all. Thank you very much, Chong. You highlight there the challenges of working across users and authorities, and I think that's going to be, you know, analogous, very similar to the challenges we're facing about working across ecologists and IT professionals and engineers and so on. And the importance of spatial management in fisheries is analogous or very similar to any other type of control or any other type of maintenance rules that you might have. And it opens the question to me of what type of themes might we need to consider to help ensure our work that we're doing is ticks off the governance box. So we've listened today about many people speaking about the need for the right type of labelled imagery. We've also, you know, spoken about connecting different types of communities together to realize what we'd like to realize. But we potentially need a theme also to think about the rules, the rules of the game, whether you're going to call that data control, privacy, access and so on. And I think this is going to be a bit like how they start to think about self-driving cars. What are the rules for these types of things to operate on a road? What are what is the rules that we'll have to think about once we start to collect data sets that are informing the management of a fishery or an agriculture venture. And that data is being used by others. So for example, we heard in the last talk about how that data held by the Nature Conservancy is open for collaboration, but it's not open yet. And how do you foresee some of the questions of of sensitivity around this information and what kind of themes questions have come up around that thinking about your app? Yeah, great question. Thank you, Mr. Kim. And as for the perspective of our NGO, we think that NGO mainly plays the role of interconnecting between the government and the people. And by that, we think that the government only do what they needed to do. For example, law enforcement, which that we cannot take on by ourselves. But as for the others, like remote censoring or data collection, or even comes out with reports that can all be done by the academia and the local communities. We all should do what we do best. And as for the government, they are great with powerful legislative measures. They are great with enforcing stuff, for example, to turn your fishing ban. But when it's come down to scientific researches and local community protections, we think it would be the best if you left them to the one who does it best. I don't know if that is a question, but I think that's what the NGOs do, that we optimize each other's job. If a local community wants to protect their water areas, then we let them do it with all the governmental interferences. But if there are things that need to be done legislatively, we will seek governmental help. Yeah, I think that's it. Thank you, Matt. Yeah, thanks so much, Song Beo Chan, for your hard work over the past days putting that together. I think it's important to note you didn't touch on any presentation, but you are using fish classification algorithms in those processes, I think, aren't you? And I think another thing from your presentation, well, it's kind of already been said, but congratulations on joining the dots between actual enforcement and empowering good management, and actually doing the collection of data as well. Because I know, for example, there are many, many projects around the world that use spatial data, that use AI data, that use social data, like the environmental witness app that I mentioned earlier, or Karatagan patrol for the Philippines. And while many of these projects are aware of what's going on, it's quite rare to see a full workflow from reinforcing to enforcing, actually. And I think that's a real massive achievement that we can all learn from. And congratulations on doing it. It's a great step. Thank you.