Select Global Data Pledge Launch – A foundation for tomorrow
8:02AM Jul 17, 2020
PP Good morning. Good afternoon. Good evening. And welcome to the AI for Good Global Summit always on all year long. We hope that you, your family, your friends, and your colleagues are all keeping healthy and safe. My name is Fred Warner from the ICU. And it's a privilege for me to introduce today's webinar. Now, vi t u is the United Nations specialized agency for information and communication technologies. And we're also the organizers of the AI for Good Global Summit alongside XPrize Foundation, in partnership with 36 un sister agencies, ACM and co convene by Switzerland. And the goal of the AI for Good Global Summit is to identify practical applications of AI to advance the Sustainable Development Goals and scale the solutions for global impact and like much of the world The AI Summit has gone digital. And we're moving forward with weekly programming, allowing us to reach more people than ever before. Now, one of our claims to fame is that we're action oriented. And we're not just a talk shop. And I think today is a perfect example of that. Earlier this year, we created and launched the Global Initiative for AI and Data Commons, which aims to create an enabling platform and infrastructure so that we can scale AI for Good problem solving. And today we're launching one of the new projects from the global initiative called the global data pledge. Now before I introduce today's speaker, I'd like to go over a few housekeeping issues. First of all, your microphone has been disabled, so please use the q&a and chat function if you wish to communicate. We'll also be running some live polls. So keep keep an eye open and when you see a live poll, please take the time to answer so we can collect your interest. And it's the job of the moderator to identify and ask questions to the panelists. And we're counting on you to have a very interactive session. And speaking of being interactive, challenge for you. Can you please let us know where you're connecting from? Simply use the chat function and type in what country or city or wherever you're connecting from and make sure you enable the chat to everyone and not just the panelists. Here I'll go first Geneva, Switzerland, India, USA, Montreal, Scotland, Georgia, Washington, DC, Madagascar or Rome. Wow. This is really truly a Global Initiative, global audience. And it's actually quite fitting, because we're launching a global data pledge. So I think we have the right audience here on hand. So without further ado, I'd like to introduce today's speaker, first speaker, and his name is Richard St. Pierre, and he's the executive vice chairman of C two. And he's also one of the driving forces behind the global data pledge. But first I'd like to show a quick video genuine please
Thank you Frederick and thank you for the Ipu for hosting us today for this session. Welcome everyone. One from Tokyo to Mongolia I just to Montreal, my hometown, and thank him for the time for joining us today. on this call, there's likely not a single person that has not been affected by dependent making one way or another. We all know someone that was affected or even lost a dear one. And if that's the case, our hearts goes up to you. There's clearly no going back. But we need to create a new normal and that new normal does not have to be all negative. That's why we're launching the global data pledge. This is not another competing Open Data Source encryption about storage and privacy laws. It's a ledge. So today, what we would like to do is define what is that global data pledge. So we're going to present to you with through our speakers and through our agenda, what the pledge is But also and more importantly, we like, like to have your help. We want you to be involved into the definition, the design, the roadmap of that pledge, so that we're not just a Me too, but we are an enabler of what already exists. We want to leverage what already exist and be a catalyst for that. So it can serve the common good. But we go before we go to the pledge itself. I'd like to introduce you a few of the speakers that will share the stage with us today. So first of all, illustrious Muhammad Yunus, Nobel Peace Prize Laureate, I'll come back to him in a few seconds. And yeah, Calderon, Executive Director of open data center, Jamie Boyd, Chief Digital Officer of the government of British Columbia. As you can see Canadians are well represented here. And I'm joined to co host this session by a mere benefit me, chair of the global AI and data come in initially And also the general manager of XPrize.
So, our first speaker
Professor Yunus is joining us directly from Dhaka, Bangladesh. Thank you, professor for joining us at such a late hour.
Thank you delighted to be here.
I think that many people can look at your Wikipedia page for your bio, but your resume cover basically 10 lifetimes. Little known fact is the fact that you have 7600 degrees and counting I think. And I'm doubtful that anyone on this call can top that. And most leaders of state call upon you for your advice. So that's why we are asked you to join today. We would like your advice on a couple of things. So first one You've assembled a list of very influential people for the world over to declare COVID-19 vaccine a global common good. Considering the what we've seen as a fragmented country by country response for this pandemic, you think that some type of data should be also labeled a common good?
Well, there's some data already common good. For example, you mentioned vaccine. Before the vaccine process started, scientists shared information with each other globally. And then starting point is laid down. So these are common good, nobody's charging for that. nobody's asking for a particular fee for that. So there's lots of those information available in the open source data with many internet you go your habit. So there are lots of alternative vailable. But at the same time there, lots of information is locked up. If you just looked at the pattern rights all over the world, people don't use it anymore. At one time, it was important. They had the patent. But nobody unpatented them anymore. So they're locked up. And it's a tremendous resource. I said, Why are you guarding them that you're not using them, this should be criminal to do that, because somebody can use that information to do something else. So that's what there is open source. But at the same time, there's a lot of information, locked up data, many, many places, that's what we should be doing attention to.
That's a good, that's a good action point. You You wrote an article, title, no going back and it made the front page of many newspaper around the world. you currently have like north of 40 million clients right now. So lots of data, what is in your mind to keep opportunity to gather knowledge from all this without Of course infringing like privacy laws or privacy rights or anything of the sort, but we all sit on the massive amount of data that you say like, like you say is locked up, what what would be kind of an easy path to that you would suggest to unlock this potential?
Well, data is knowledge, that data is a bagel through which you share knowledge. So, that knowledge should be shared this is very simple formula. So, this is this is something that we keep on promoting that it can be done, but the problem of the data is not all data are not harmless. Some data in the shape of data, the contaminated things, he will destroy the whole validity of the database itself, and some are manipulated. You go quietly inside of it and manipulate your minds and so on, so forth. So Same thing that we have to guard against what is the what is the right data and who is the wrong data. There are lots of people waiting around to manipulate data. Like in elections you're talking about, like you're watching television, and then your mind is manipulated. You give out information, fake information. Fake data is galore. So, these are the troubles for that even when the whole world the whole science of statistics came. It was started by saying lies, damn lies in the statistics. And that's where this all comes from. So, while data is important, at the same time, we have to be extremely cautious that how to retain the validity of the data, how to retain the purity of data, that because if there are so many sources by which they are being manipulated to find a way to make money or Because some political security advantage out of it. So this is what we should be guarding against.
Before I turn to the audience to maybe ask a question, and if you want to ask a question to Professor Yunus just drop it into the chat, and we'll pick one for the next question. But I live for I live what you talk about manipulation of data when I was last in Bangladesh with you, where we could see that how the media portrayed the situation really gave a negative spin on something that was very positive, the conference that we were having together. Considering we're talking to have a data pledge, we're not going to say to everyone keep your data and don't do anything for it because it might be misused. So what are the key things that you think that what's the message for the critical thing that we should look for, to make sure that data is used in the proper and that serves humanity itself.
I thought that when you mentioned data pledge, I thought pledge is a very mild word. It depends on your wish, you want to clip it and you want to break it. Nobody kind of punish you for that. So it should be harsher than that. It seems to be. If you contaminate if you manipulate if you distort data, it should be a criminal offense. So then we understand what we're talking about, pledge, just a nicety sign, then you go home to do whatever you want, nobody is going to punish you, nobody's going to take care of. So I would recommend that we get harsher on that. So that it's not something that you play around with life of people life of the future life of the planet itself, in manipulating and twisting and distorting the data.
I think Paolo follows one of the audience member follows to what you were saying, Do you think we can go all the way to oblige or made it mandatory that some data be shared, that might not be received very well by many people, from individuals to countries?
Oh, I was saying isn't just a patent dragged open source and patent rights kind of thing. So I would say instead of patent right, it should be patent left. Instead of holding it, you turn it around, make it available to everybody. That's the way to proceed. So that that's the kind of thing that we should be looking forward at the same time guarding the purity of the information, purity of the data. So data is not be locked up in any way, for the advantage for differential advantage of people for security. I understand that there's some security issues something but beyond that, there is no reason why we have to lock up those information and deprive the human MCP to take advantage of it.
I think Angelica from Toronto summarizes well we should start by data manifesto of some kind and then move on to more like, harsher measure if need be. But that's all the time we have Professor Yunus. I thank you very much for your time. And we'll I'll now pass it on to Amir. For our next speaker. I'm here. It's all yours. Thank
you. Thank you very much.
Thank you very much, Professor Yunus. And if you're still with us in a few minutes, we'll probably ask more questions to you. So welcome, everyone again, and thank you very much for joining us and thank you, Richard, for opening the session and welcoming us with Professor Yunus. I would like to introduce our next speaker, Anya Calderon, who she is the executive director of open data charter and she essentially reformers from around the world to implement Open Data policies based on shared principles. And she has successfully also delivered to keep presidential mandate for open data and Mexico prior to that, and we have to have a few questions for Anya. But before I ask you the question Anya as a context, for years ago, we started this whole movement called AI for Good and in magnificent bay by a series of summits and gathering. And what we discovered is that access to data is not obviously evidence, there are many Open Data everywhere, and there are private data as well. And when you think about agency of problem solving, and given people that are trying to solve local problems, the ability to address those problems, it was not evident. So we started talking about the fact that AI and data should become a common good, how is the is the right question. We don't know how but we're trying to make sure that we link to for the common good, and that starts with basically access to some information and data. So from that perspective, I would like to welcome you and ask you the first question is because you have been the origin of this open data charter, maybe can tell us more about your charter, but also, what should be looking at some impact metrics when we talk about open data.
Thank you, Amir. And first, I want to thank the organizers for the invitation. It's a pleasure to be here today. As mentioned, I had a global organization called the Open Data charter whose goal is very, it's at the heart of the Open Data charter to make data available to people. It's a collaboration of over 100 governments and expert organizations working to collect, share and use data to address targeted policy challenges. And we do this by developing practical guidance to open update and waste that Deliver tangible benefits to people. We believe that this can only be achieved by governing data, inclusively and accountably. Now, to your question, data has often been presented as a vaccine for the perils of human Cotton's from revolutionizing sustainable development goals to fueling an equitable pandemic recovery. Today, it is hard to understand a world without data. And I've always thought of data. I've thought about data as something that can inform better decisions. And this can lead to value in many different ways. For example, if the data that underpins decision making by people and powers made open, such as from government officials or large corporations, it can sustain a healthier public debate based on a shared reality. That is what we ultimately need to improve trust and contribute to A well functioning democratic system, but opening data in ways that make it accessible, comparable and timely, which are principles of the Open Data charter can for example, in our current context, allow people to know where to go to buy masks and how much stock is available such as is the case in Taiwan, find your nearest COVID-19 testing center such as here in the UK, or help health authorities mitigate the spread of the disease using mobility data that is has been released by companies such as Google and Apple. But much like a vaccine, placing good quality data in the hands of people that need it requires a governance process that aims to maximize access to benefits while reducing unwelcome side effects. as has been mentioned before, I think what we are seeing is that is is as important for the public To be rooted in principles of openness as much as the data that has been made available through it. So this means transparent and honest communication about the data that's being published, how it's structured, who made those decisions and what limitations Do we need to be aware of. We also need to do more in terms of explaining how sensitive and personal information is being safeguarded, publishing any privacy or other risk impact assessments in timely ways. It is very much about a process that meaningfully allows those that may be affected, or watchdog organizations the ability to scrutinize to raise concerns and to encourage people to engage with that data in order to create value
is very insightful.
As to talk about value, we try to capture value today from the SDGs perspective. Because as a framework to evaluate problems or problems that needs to be solved, if you think about problems as, as the recipient of data, and we had this conversation together before, how do you think that more open data and data sharing would make sense if we knew in advance what problems needed to be solved?
Yes, I think this is a very important question because for any data initiative, context really matters. We don't treat health data the same way as we do transport data or data from the judiciary system. For example, at the Open Data charter, we advocate for this by calling for governments to publish with purpose. And this is because they're always different levels of risk and bolt. Those that are deemed low risks should perhaps be open by default, while others require more degrees of safeguarding people may feel comfortable with certain types of data to be sure to share for certain purposes. This and not others, and in cases that involve data about, for example, those people in positions of powers and expenses of you, your commissioners or other public officials, there are laws that require that information to be made available in order to redress that power. And fundamentally, this is important because as humans, we experience the world through problems and not through data. In other words, you first need to clearly articulate a problem that resonates to people, and then we can figure out what and how can data be used to make a difference in their lives? Now, I think it's important to note a recent report in nature that points to a common focus from our industry, the AI and tech industry to ask whether their tools are fair or for good. And there have fair further claims that these are, tend to be infinite spacious words, pushing us in instead to be asking questions like How is data or in this case, in this case? Well, the article focuses on AI. But in this case, we can ask ourselves how data is shifting power. So it calls for putting marginalized and impact communities at the center of design, development and governance of these processes. Along with thinking about problems. I think it's also critical to position those problems as part of a political economy that acknowledges power imbalances, especially as the pandemic surveillance and data gathering efforts are accelerating the risks of equity implications that may potentially be hard to roll back. So finally, with with the World Health Organization, now calling for rapid data sharing as the basis for public health action, we really need to ask ourselves, how can this tilt the playing field to benefit those that are most likely to be impacted or incriminated against
a lot of your your, your your positioning on diverse, vulnerable communities.
Can we say now that we need more open data today than any other time? Is it a crisis a revealing factor four more open data? Or do you think that open data should be more weaved into the fabric of our society and governance and policymaking and crisis should not be affecting that.
And I definitely am a believer of crisis bringing a transfer opportunity. So I think it's also very particular
world we're living we're experiencing a global collision of crises caused by Coronavirus, the aftermath of recession looming climate emergency. On one hand, we see how crucial it has been to open and share data in order to To help us navigate this pandemic from sharing scientific discoveries, tracking the spread of disease health sector capacity to informing policies and reforms that aim to address the fault lines that have been exposed in our political and social balance. Most visibly, I think the lack of protection for the poorest or racial divides or investments in or lack of investment in health care systems. But what has also been exposed is a very fragmented data ecosystem that makes collaboration extremely difficult. And the global nature of this problem and these crises require joined up action. So in our efforts to map we're currently mapping types of data that different countries are making available. We have seen massive inconsistencies in how data is collected, defined and aggregated, resulting in attributes that can't be compared and leaving some significant gaps. For example, Efforts to capture age, race, ethnicity and location that are very important demographic data, we need to understand the impacts of the pandemic. Some countries have disaggregated data to dangerous levels, and others are showing very good and innovative data management practices. So hungry, for example, published a list of Coronavirus deaths to the level that allowed disease to be re identified, including very sensitive information about them which violates their privacy. While for example, in New Zealand, we saw carefully managed and dynamic data practices that waited until the number of cases was large enough before they started disaggregating information in safe ways at first not releasing details, details such as age, then including age brackets and later age numbers. So as more countries are starting to include this level of detail, there's Still a stark lack of standardization. Just say the same example of age bands, and how it differs across countries, some use ranges from 30 to 39, others 35 to 44, etc. It makes it very hard to compare these numbers. So in an increasingly polarized and challenging geopolitical world that we're living today, I believe that opening and sharing data across borders in ways that are enhancing collaboration efforts is a vital step that we need to take in order to make progress together.
Thank you so much. And yeah,
there are a few questions. I think some of them are for you in the q&a, let you look at them. I'm going to go to Jamie and I'm going to come back to you. So we'll actually introduce Jamie Boyd. I'm very pleased to have Jenny with us today. She serves as the Chief Digital Officer of the government of British Columbia. Canada and she does lead efforts to embrace digital change, helping governments serve citizens using modern technology. And she's a big advocate of government transparency, accountability, and the common good. Jamie, welcome. And we're so happy to have you. And we heard the answers from Professor Yunus and also on your Calderon, you have been talking a lot about investing in open data. And then maybe can explain a bit more what you mean by investing in open data? And how would you translate a pledge of data as an investment?
Absolutely. Hello, everybody. It's a pleasure to be joining you today from the traditional territory of silicon wood speaking people on the west coast of Canada. This is such an important and exciting topic to be discussing with you all. I have spoken at length about the value of investing in open data. And I think that that's the right characterization because Open Data data is not just a thing that happens Releasing and using data is never neutral. And it's very expensive. It's very cost intensive in terms of labor and whatnot. And I think that that's a really important frame to bring to these kinds of conversations. Sometimes we think of data as something that just flows around us and, and is just available for us to pluck out of the sky. It really isn't. It can be extremely challenging and costly to manage these things appropriately, thoughtfully, and in a way that is really citizen driven or human driven. And so the perspective that I bring to this is one that's very much from the public sector. I've spent a number of years working in government. And for us, our our guiding north, the reason governments exist is to serve people. Data is a powerful enabler for that kind of citizen driven service. And given that, it would seem to me really useful to to frame data collection and use as being something that can and should be human centric. At the same time balancing the imperative of leveraging data for service delivery, with all of the costs that come with it is is, you know, it can be challenging the kinds of things that in my mind, make it somewhat costly to effectively manage and use data. Or that well, it's, it's a question of availability. So often, large organizations like my own will collect data. And, and we do need to be somewhat nuanced in the way that we triage the kind of data that it would be a candidate for release. So as Anya quite rightly said, we need to be thoughtful and publish with purpose, right? It's not reasonable or realistic for larger organizations to simply put all content out without any thought to the availability to find ability, the extent to which it conforms to data standards and whatnot. I think that in the brave new world of the fourth industrial revolution, and dramatic innovations, it's also important to think about the way that data flows. So not just putting out a data set and saying, okay, we're transparent now. Right, you can't have a one time investment, these things have to be refreshed, they have to be maintained, you have to be constantly assessing the data that you make available with that that's available from third parties. Because we do have the risk of re identification. And many of us are guided by robust privacy frameworks and for good reason. And so in that kind of an environment, we need to be thinking about the availability of the data, the accessibility of the data, are people going to be able to find it, just because you released it as an organization doesn't mean that the person who needs to be using that data is going to be able to find it. And can they use it, and the the time and the hours and frankly, the lifetimes that we've spent cleaning data is not great value, right? And so we need to be aligning with this. Establish data standards. And finally, we need to be thinking about Lifecycle Management. So not just spitting out that data one time and declaring victory, but really having this ongoing dynamic. So yeah, I would say it's an investment. It's an important one though.
So if we talk about data pledge, in the world of investment, when think about the investment, also nurture that investment, we just don't invest, we try to accompany is right to make room for it. Give him a chance to succeed and to grow. So if if a data pledge into open data or in general during open, which is a is an investment, who are going to be the investors, are we talking about just an organization that are government, NGOs organization who can pledge data, or is is my little nonprofit in in Kuala Lumpur can be a Giving Pledge or should we be government? Who are the best candidates to pledge their
That's a great question. And I mean, I'm almost going to cop out of this one, because it seems to me that we do need to be taking an ecosystem approach. So the data owners, regardless of where they are in the ecosystem, really should be thinking about whether the data is going to contribute to common good. And if it's going to be able to nurture these kinds of ecosystems, I think we do need to recognize that there are lots of reasons why organizations might not want to default to opening all of their data. So the privacy considerations that I identified earlier, many organizations are looking at, you know, their competitive advantage being driven by some kind of buy buy at the data they manage. And those are important considerations. At the same time, a lot of very data rich organizations can nurture ecosystems and build innovation around themselves by considering, excuse me, considering things like releasing data aggregates, so releasing a portion of the data in a way that would inspire it Innovation or enable other service providers to build on top of the data offerings, while also respecting privacy or you know, competitive dynamics, whatever they might be. In the public sector, of course, we also have security considerations and all that sort of thing. So I would really say the entire ecosystem, increasingly, organizations that are not governments are becoming very data rich. I think that when some of us got involved in the open data movement, certainly from the public sector, in my case, we were sitting on massive amounts of data in government, and it was a question of going out finding that data, cleaning it up and releasing it. And we were able to make very significant progress in my case within the government of Canada. And increasingly, though, the kinds of data that are available from private sector organizations are going to be significant. Just super anecdotal, but I think it's helpful you Think of road network data. Who would be the best provider of road network data? Well, you sort of instinctually you'd say government makes sense. But now with the fleet that would be managed by an entity like Uber or Lyft. real time data for road networks is going to be available through that kind of a data source. Right. And so if you have sort of trusted sources of truth coming in from the public sector, you can then take cross sector approaches, working between sectors to be very effective. I started my remarks by saying that data is not neutral. So I would also argue that a really important set of players in this ecosystem would be academia but not for profit sector. And then whoever it is that is releasing this data, managing it, ensuring that we are supporting inclusion and effective service delivery, there is going to be an ongoing need for for capacity building. Because again, the data that that many of us I have the privilege of working with does have fairly intimate implications for people's well being.
I'd like to come back to, to question to you and both Anya about you mentioned the the concept of common good. And once we launched this past January, the Global Initiative on AI and data for the common good. The notion of common good is not an obvious one, because we all care about a common good, but one becomes a public utility and usage he has to be available and sustainable to remain a common good. So as we talk about open data, you believe that it pledge of some sort could be a starting point, to engage into the understanding that the common good needs to exist.
Thank you, Jamie.
I think thinking about that objective is going to be very important. For what success looks like for a global data pledge, because you can only achieve data for the comments if it's developed by the comments, not just for the comments. So I think thinking about, you know, what our communities and individuals that that data is intended to serve. I mean, what we discussed before clearly articulating that purpose and those problems that you're, you're seeking to, to find solutions towards, it's about involving people in that process of understanding, you know, what levels of openness are they comfortable with in terms of sharing that data for those purposes, and also making it about continuous process because as you're making data available, and Jamie will know things Data is never perfect. It's about a continuous improvement process. Once you start using that data, you will understand how it can be improved, but also the world change norms change. people's beliefs and values may change over time. And we need to continue to bring them in the data journey and governance process to make sure that it is serving that common purpose.
Yeah. So I think this is a great question. How do we make sure the data lives up to its promise? And I think that the answer, a good answer to this is one that really looks at data as being the lifeblood of so much of the service delivery that we have across sectors in a digital world. So I again, have the the pleasure of working in a government that is very focused on providing excellent services to the people of British Columbia and we talked about Best Practices in service delivery. And I think that those best practices and in developing digital services are best practices that also apply to the management of data and making it available. So those best practices are things like being agile, human, centered and open. So when we talk about being agile in software development, it's really about iterating, about setting an objective, taking a few steps towards it, learning as we go. And there are lots of different strategies for doing that in the data space. So you know, making for example available the whole suite of data that might be candidates for release, and then working with communities to understand what would be highest value. This is something that we've seen in our movement in the open data movement over the over the past few years, and we've learned a lot. We've learned, for example, that the data standards that we had for gender and for gender were not appropriate. And we learned that we had to adopt non binary data standards. We've learned as we go because we work with communities and the Communities tell us when the data that we are providing is or isn't appropriate. We can be human centered. So for us in software development, that means watching people interact with the services, understanding their journey through the kind of problem space that we're trying to address. That's a similar kind of thing that we can lend to the data space, inviting the people who are going to make use of the data to explain what the use cases and really living the ethos of publishing with purpose, and finally we can be open. So really defaulting to to open standards. You know, we in in British Columbia were to the extent possible in open source government. We like to work with communities, because it makes things better. We're able to develop those robust ecosystems were able to collaborate and ultimately that that leads to better services. I think that too often in our in our data driven society, we feel like Data is a thing that happens to us. And I would argue that there's a real opportunity right now, for us as citizens of the world to come together and say, No, we are our own data owners. We believe in things like self sovereign identity, we believe in our ability to actually manage our own our own data, and have a view sophisticated view on who and who should and shouldn't be accessing that data or making use of it. So it seems to me that we need to drive ourselves to a point where this vision of citizen centered or human centered data is no longer revolutionary, because today it's a little bit too revolutionary.
On the phone, thank you.
We will have time we get two more questions. There are many questions out there but
these are the principles that are driving us all together as a community and well there are sort of principles like open data charter has created principles where there are European Union's have a number of research, but many governments like yours British Columbia government has already adopted a set of standards and principles and guidance. And there are numerous think tank or research organizations that are alerting us about privacy, safety, security bias and so forth. How can we incentivize more data sharing, and incentivize more responsible, right pledge and investment of data for the common good. So that's the goal of this pledge that we're trying to to, to start with and Richard is going to be explaining how we're going to be how we're planning to do this, see if we can have the next slide.
And we'll come back to you.
So the goal of the global template is basically saying that, of course it is our collective mission to find ways to identify and manage and distribute data in order to respond to all crisis whether this is this pandemic or hunger or Poverty, or gender inequality, there are multiple crises around the US that come and go in importance. And the goal for us is to propose an intentional pledge on global data resources. We're not trying to recreate all the effort that already exists. There are charters, the presence of Ania. Here is a proof of that is that we're going to be relying on those existing charters. And so the principles, but how can we incentivize? Now that we have those principles? How can we incentivize for more data sharing more data pleasure, so we can learn more, we can share more, we can collectively solve problems together more. And in order to do that we are we're trying to announce this at the AI for Good Summit is September. But in the meantime, we'll try to create two working groups and reachable by explain them shortly on how we can let you go through that how can design a playbook of data pledge and how we can increase collaboration between all the parties and spinning up right access data, which are up to you.
And me, we thought that data was old news. We've been talking about it for like decades and here we are. It's still centerstage. So data is the crux of what we need to do. I love when Anya Calderon says we need to have a defined a problem that resonates with people. And so the two work groups that we're initiating are exactly for that. So it's not going to come out of my head alone. Amir's head alone, Jamie's head alone. It's the collective that needs to define the framework to address those two questions that you have there on your screen. So we've there's many ways to skin a cat. But these are the first initial steps that we need to take in order to create kind of a reference point in order to move forward. So the two the two work groups that we are planning our first data pledge playbook design, and the second one, collaborate And scaling and be ready you in the audience, we're going to be asking for your help on both. So let's go to the first data pledge playbook design. So first, it's a pledge, but what does it pledge look like? And we heard from Muhammad Yunus that we he wanted to be more incisive, then pledge. Let's see how we can make it or what is the first step that needs to take it. But let's be clear, it's not a storage approach. It's not about encryption. And it's not about the new. redefining privacy laws. Of course, we need to address those. But we need first and foremost to understand what is out there and under which conditions these will be pledged. So it we're talking here about a framing document. And we're also talking about under which standard which protocol, it would be shared and so on. So these are the key elements of what we would like to have as the data pledge playbook designed. And ultimately, as I mean was saying by end of September, we would like to have the first result of that definition to be able to come back to all of you with your collaboration and say, what does that look like? So I'm gonna ask the first question, do you want to be part of that workgroup? And you're gonna have a pop up on your screen right now? And if you want to be so, click on Yes. If not, click on No. And even if you want to be one of the leaders of this group, click on the button as such.
So Tick Tock 15 seconds, please answer your question.
And as you're answering the question, we will be reaching out to you by email in the coming days in order to draw up the plan and the roadmap between now and September.
gang, I think that was enough time for just one click. So we'll move on to work group two. So we're group two, we've labeled collaboration and scaling. So our goal and we've said it multiple times is not about reinventing the wheel. So we have to figure out the activities that will enhance and amplify what already exists. How can we act as a catalyst to what's already there? There's more than 1600 initiative worldwide from open data to other standards. How do we make sure that worldwide we have a global footprint and it also means inclusiveness here, it's not just about the US, France, Germany, we saw that some people are from Iran on this group. So we want to be very inclusive and include a lot of countries. So please answer the question as to collaboration in scaling. Are you do you want to be part of the group or not?
Again, I think that was ample time for people to just do a single click. I thank you very much I look forward to reaching out to you over the coming days. Look at my email address over here if you want to exchange in the meantime, if you were unsure about your position on that, I'll be happy to exchange with you. In the meantime, I'll pass it on again to Amir to show and tell us a bit about next steps.
Thank you very much, Richard. And I hope these two groups were were simple enough. We tried hard to be really basic in the approach and not make sure we recreate the underworld again. So as Richard said, we're going to be reaching out to all volunteers that want to join us in this working groups are going to be short lived. The goal for us is to have a first designed for shop as soon as next week. And then during the AI for Good Summit, which are From September 21, to 30th, we'll have probably one or multiple reports out during that time about what happened, what we learned and how we can actually start to pledge. And this is an important time I think pandemic has shown that we can get together, we can come together, we can work together. And it's probably also time as both Jamie and Anya said that the narrative of data is no, we're not learning something new. But the way we act and participate, I think should be reminded on a constant basis and we want to have this pledge as a signification that we care and we want to basically give all the problem solvers or the problem owners a fuel, right an engine to connect them and data is one of them. That's not enough. We need many more we need governance when it many other other ingredients, but this is an important one, if time allows and will thank you all for participation and comment. We're very excited that we've joined reach out to every one of you. If we still have time, there are a number of questions and if Anya and Jamie are willing to, you can select the questions that you want to answer and answer them directly. Live on this on this panel.
I can read them to you if you want to.
Or you can select the one that you think you want to answer a cancer now, because some of them require a longer answer.
Sure, in the interest of keeping the conversation going, and there there are two questions. That really jumped out at me as being really quite fascinating. So one is around the connection between data and AI. And I think that that's a great question. And then the other one is sort of how to, is there scope for us to be obliging data owners to open up data? So I think these are both fascinating questions. So I'll start with the second one. I don't, I don't know that that's the best approach to necessarily oblige. There are some organizations, some governments that certainly have looked at at legislating or at least legislating elements of therapy and data approach. So the there are, for example, the legal framework that came into effect in the United States last year, does look at having a legislative requirement around data stewardship. But it seems to me that there are other really interesting ways to consent, the availability of data, so doing things like running an open data charter as Anya does, having things like this pledge where we're essentially looking to orchestrate a race to the top. My previous role was in the was in the Government of Canada. And I have to say one of one of our favorite things was the International, the Open Data barometer, because it got all of our competitive side going. And every year, we would look at the results. And we're like, all right, this year, we're fifth. All right, this year, we're second, we're gonna keep working on data standards. And then last year, we tied for first with the United Kingdom. It was like all right, all of that really frustrating and time consuming work around data standards that really paid off. And so I think that really thinking about having that kind of Race to the Top can be very effective. And I would also say that demonstrating the value of the data making use of it, and telling those stories showing the impact indicators, talking about businesses getting created about services being provided. Those I think are really valuable ways of trying to encourage the further release of data. The other question that jumped out at me is around the connection. Let's see who's this, the connection between data and AI. This is something that I think that we probably need to talk about a little bit more than we perhaps have in the past. Because in a world of so much use of emerging technology, we do have to be thoughtful about how we run our algorithms, how we train them, and that sort of thing. And so in, in our government, we do make use of AI. And I would argue that to be useful AI does need to ingest a lot of data. It has to be, you know, to the extent possible data that allows for explain ability. And so when we do go out and we put out high volumes of data, it can be useful for training the algorithms, it can be useful for explained ability. replicability. So you know, to use algorithms for to support decision making, it's important that at least in the public sector, that we have the ability to explain how the algorithm works and how it led to recommendation and that sort of thing. transparency, you know, legitimacy and frankly, just ease. When we work in the open data space, there are fairly established strategies for adopting open standards. And for you know, my background is I came into government as an economist. And as I mentioned, the number of years that I've spent cleaning data is just not a good use of anybody's time. So aligning with established standards can be very helpful for accelerating some of those those innovations that we see in the emerging technology space. So those are some some great questions. I'd be really keen to hear ideas view, though, on that first question around how do we how do we encourage the release of open data? Should we be obliging folks
Thanks, Jamie. I mean, there are already laws in place that make certain data available or for example, for EU regulation that have processes in place for citizens to access data and even open data, policy policies being embedded within access to information, regulations. I know Canada has a really good job around that. So I think in some case, it yes, we do need regulation to support that. We ensure the data is make made available. But also we need to go beyond regulation that's pushing for data to be available and thinking about other types of risks and mitigation efforts needed to make sure that we are As, at the same time as pushing for advancing transparency and accountability, also making sure that we're protecting people's, both individual and communities rights along the way. And to other questions on the tact. To continue the discussion, I answered two very good ones in in the chat box about political motivations, and and what makes data risky. But I also want to point out a really interesting one around bias ease and invisible biases that may creep into data that potentially make it hard to or can become blind spots in us being able to truly address these both benefits and harms. And I think I mean in the question And you point out a very important thing, which is making sure that we're aware of those biases. But for data publishers, I would encourage that that to be clearly defined and communicated within how that data was collected and structured often, you know, I would say it's practically impossible to D bias, any data set or tool but being aware of the positionality of the people that were behind in structuring and sharing that data can help us understand, you know, where are their potential areas where it can lead to discrimination or for those those biases to produce harmful results. So yes, I would, I would completely agree that this is something we need to be more open and transparent about and also proactively putting efforts in place to make sure that we're mitigating those risks.
It has been wonderful, thank you so much for for your answers that are so rich and the depth. There are too many questions, we will have time to go forward. But there are a couple of questions that talks about politics that we may not answer today. But I think part of the working group need to consider those questions about the fact that there is a lot of politics around AI. So the fact that in terms of global powers can basically violence sovereignty of some countries, and we do need a data management convention under the UN. So probably, this is not ours on the pledge site to respond to that we're trying to promote right pledging more. But I think both of you in your respective networks, these are something very important. We try to surface them as part of the working groups and refer that I probably mentioned them during the AI for Good Summit, I think these are important questions of sovereignty. And it goes from personal sovereignty to community service to the country's sovereignty. And those answers are difficult answers. I would like to thank all of you for joining us today. I think we're coming to time. And on the end, Jamie, it was wonderful to have you. Thank you so much for for coming and sharing some of the insight and the knowledge that you have gathered, and we're looking forward to working with with you on an ongoing basis. And this is the very start of that, and we're looking to get ready for September.
Again, I'm asking Fred to to help us close this session and get ready for this one. Thank you. Oh,
thank you very much, Amir, and I'd like to sincerely thank all of the panelists and speakers for such a great discussion. And also all the participants of course, both on the zoom call and on social media who have been following the live stream. Again, from the air team a big big thank you Just a few housekeeping issues. So we have another action packed week next week, we have three webinars in a row. The first one will be a new breakthrough session called the collective pandemic Alliance. And that's been put together. And we'll be running that on Tuesday afternoon. And also on Thursday, we'll have part two of our global dialogue on eSports. So if that's a topic that's of interest to you, please tune in. And last but not least on Friday, we'll be running our innovation life pitching session, where we're trying to identify startups that are using AI in a unique way to advance one or more of the Sustainable Development Goals. So a very busy week next week and all of the registration links are being posted in the chat. So please go to the chat. If one of those grab your attention. And with that, I once again want to thank all the panelists, participants speakers, but also our partners. co organizers. XPrize partners, un sister agencies, ACM co convene with Switzerland Then of course, all of our sponsors. And with that, I wish you good evening and goodbye.