Global AI Policy Landscape and What You Need to Know | ASU+GSV Summit 2024

[Music] so welcome to this important panel about global AI policy landscape and what you need to know I'm honored to be moderating this panel today and I am actually an ech founder from Europe and you might have heard about the European AI act so whatever we do in Europe when it comes to education we need to have a safety first and um data protection policy um and this is the European perspective but today we want to take a global look and learn from experts um about the global um landscape and I'm uh I will shortly uh present the the panelists today so we have Paul leas um he's beer head Global Ai and information policy initiatives at the software and information industry Association and he's advocating for responsible ech AI practices next we have ragad Alan she's an adviser to the United Arab um Emirates minister of education and she focuses on merging etic Ai and Entrepreneurship to dve drive educational Innovation and next we have man Lamar and he's Deputy assistant secretary at the US Department of Labor he Blends his experience as a former teacher with EXP expertise in Workforce Development and AI policy and last but not least we have Dean Klein uh and with a legacy inventor investing and ech Innovation Dean advises the World Bank ech team and leads Global engagement for the Sarah research Network so please a round of applause to our great panel today so and Paul I'd like to start with you and ask you to give us um your shared Opera operating knowledge what should we know about the current Global AI policy landscape concerning education great thank you for that introduction um thank you all for for being here today so uh over the past several years has been an explosion in policy globally related to artificial intelligence um over 50 countries have develop national strategies on how to harness and develop AI in a responsible manner and according to the oecd there have been over 700 initiatives in the past 6 years alone around AI launched by uh National governments um I would note that only five of those are in Africa and that's something I think we may talk about a little bit today um regarding education it's a very different landscape um so generally around AI a lot of people know about the European artificial intelligence act which um is going into Force this year um or has been passed and will go into Force um over a staggered schedule over the next two years um there have been a lot of uh actually pieces of policy and regulation that have been enacted in other countries um but overall I think in the education space countries are taking more of a wait and see approach one thing we've seen quite a bit is the development of principles and guidelines from National governments from Departments of Education Ministries of Education about how how to use AI in the classroom uh but there's a lot of impediments to actually doing that and sometimes it comes down to not having the data not having the tools uh not having the infrastructure that you actually need to deploy AI in a responsible and ethical way um so uh I will say a bit more about that as we go on but um just wanted to give an overview here uh one further point before we move on um I me Africa uh when we're thinking about education we think about um the the potential to really help um in in countries that may not have the resources right now and that includes a number of countries in Africa the African Union development agency issued a report in February that outlines a pretty ambitious road map for adopting AI uh more broadly not specific to education that is expected to be voted on in a year uh but there's a great deficit not only in countries in a in Africa but in a number of other countries um and we are quite a ways away from actually adopting AI in education and in other spaces uh for example the United States over 90% internet P penetration in the United States is at over 90% in subsaharan Africa it's in the low3s um so we we still have to get that infrastructure in place before we can talk about unleashing some of the amazing potential that AI can bring in the education space thank you so much Paul so rad how is the situation uh in UAE on the one hand like AI in schools but also on the policy level um so AI is already transforming the way we live the way we work and the way we learn and interact with each other um we in the UAE have have realized this um back in in 2016 when the UAE assigned its first minister of AI a year later in 2017 we introduced the national strategy for um for AI that included education but if I would speak specifically about the education sector we can see that the conversations have shifted from research and Academia in education to the Practical development of education Technologies and education tools that result resulted in an ech U Market size of around 140 billion as a global ech uh Market size in 20 23 with an estimated growth of 13 to 14% from today in 2024 to 2030 almost doubling reaching uh 250 billion USD as a global ech market size now we as Ministry of Education as policy makers we know that if we don't Embrace AI in education we risk eliminating or alienating our formal education system from our students and therefore we made a very strategic decision as the UAE Ministry of Education um just after the release of jbt in November 2022 end of November 2022 our minister in February 2023 was on stage announcing three key uh decisions for the Ministry of Education one that we fully Embrace AI in education and two that we will be conducting a systemwide review to see and assess how we can integrate AI in education and we've started that process already when it comes to policym curriculum development um education delivery and assessment and three we launched our own National AI tutor that is Dev developed by um e tech startups in the UAE like uh ASI and Al if in partnership with Microsoft and other uh e tech uh or technology companies um we've we've embraced that uh that decision very early on although we know that a lot of countries have decided to opt out and to ban uh AI in education we decided to take that risk and to embrace AI in education uh with the initiatives and these tools and we know that they will be reflected in three key areas one is the student facing tools like the AI tutor that I talked about there is also teacher facing tools like the teaching assistance um and uh the other tools that will support our teachers to remove some of the admin burden and three is systemwide tools that will help us as policy makers to get the data and insights that we need to make better decisions when it comes to education and also to support education institutions and School principles in running their their education uh institutions more efficiently and and effectively wonderful seems like you are really far ahead uh when it comes to integrating AI into education you you've asked me about policies as well right so more specifically on policy uh we're currently working with with our partner I I'll make it I'll make it short so we're currently working with different Partners like the world economic forum and strategy and to developing our own policy guidelines and drafting our policies when it comes to embracing AI in education but in a nutshell if I would summarize how our approach is when it comes to policymaking uh we're doing two things one we're working with Educators Frontline educate Educators like teachers uh School principles and students and parents as well to draft these policies and two we're taking a very agile approach because we know and we accept that we can't get the right answer from the first goal it's a a rapidly evolving field so we know that we have to be very agile when it comes to policym wonderful thank you so much so Dean the question for you is how does the World Bank see AI uh from an investment perspective can you tell us a little bit about that yeah so I'll I'll I I often make the mistake of assuming that people understand what the World Bank is so just just very briefly um it's a multilateral organization it is a bank but the shareholders are countries so the governments and we only uh lend to or or um uh make grants to government so in the case of Education it's largely ministries of education and the World Bank is the largest funer of Education in lower and mid Middle inome Country so depending on the year it could be three to six uh billion dollar and um it's you know with respect to to technology when I returned three years ago at the beginning of the um of of the pandemic about 30% of those projects had a technology element and now it's it's 90% um so so it's put some stress on the kind of resources the kind of human capital resources of the bank and um I I think another thing to remember too is that the the time from initiating a project and those projects do belong to the governments at the end of the day they will borrow or they may get a grant for it so with education projects probably about a half of it uh are concessional that means means it's it's very low interest rate it's a grant it's a it's a guarantee and of those uh of of that amount probably about two-thirds goes to Africa um so and I think that's uh that's a rightful Focus so that that's kind of the the the background the situation there's been there's been a lot of demand um for AI uh by most governments and um we don't assume that um they understand sort of the implications I'm not even so sure we do you know most of the time so then it's a matter of getting expertise to that um you know how do you you know what sort of AI do you want and usually you know then there's you know kind of an educational process there because there's 17 types at least right now I believe you know how do you want to use it why do you want to use it it um but there's a there's a general concern that um these countries will within these countries and and by the World Bank that that they'll fall further and further behind um on AI and education and skills and you've got tremendous pressure on the part of young people um who will need to be properly educated uh who probably are not I think it's 50% of of 10year olds in lower middle inome countries can understand a simple paragraph and and read it and in a lot of the countries that get concessional lending from the World Bank it's more like 80% so it's a huge and it's in my View kind of the challenge for our time because you're having these tremendous demographic pressures too and I think by 20 50 um a third of the world's youth so think of that as like 15 to 24 years old will be in Africa uh and right now uh unemployment I think uh one out of four uh people who are unemployed or youth so we expect that to get worse too so you can see there's a lot of um concern about kind of the political economic pressures that that are going to come on on governments and you know making sure that these uh young people have um sort of you know work that that's that's meaningful to them so with respect to the World Bank we've just started doing um putting AI into education projects and that's almost entirely been in higher education um one of the concerns of course we would have are is the data security and data sharing and um in terms of that sharing that data belongs to the count so it's up to them what to do with it um we the the World Bank cannot cannot dictate that most of those countries do not want it shared not even anonymized not even aggregate data um you we we carry it but we can't necessarily you know give it out um I think there's there's also a concern about uh the policy environment whether it's correct and that policy environment is going to just keep moving so I'm thinking you know how do we use AI to create AI policy and that's something that um you know we have to work on it's kind of very very nent right now and um so I I think that kind of demand is and the situation that's that's actually pretty dire is kind of what I'd most like to impress upon you thanks for that Insight in all these numbers um so now I would like to have a really Hands-On question start with you many you're a former educator you're a former classroom teacher if you were still teaching today what would you do what would be your approach um yeah good question do I have any former Educators in the audience or current educator awesome excellent um so hopefully at the minimum you all will be able to relate to some of this so I think two things come to mind when when I think about that on the one hand it's what I would do differently and on the other hand it's what I think I would already be doing but wanting to accelerate on that on the one hand of what I would do differently um I would emphasize a lot more on digital literacy so I taught Elementary in middle school and I think sometimes us as adults sometimes we assume that if a student knows how to turn on a iPad a laptop or even use chat GPT that mean they're actually digitally literate and that's actually not the case there's two like there's two components to that there's do you know how to use the tool and so I have twoy old nephews I swear to you it's like they came out the womb doing this on screens they're like doing this they want to and so that's one thing um but actually understanding if you understand the key components of it and how like so for example another example I have um you know teenage nephews that could upload an essay and get you a good response in chat GPT but if I have a dialogue with them to actually push them to internalize what is it that this topic is about they'll struggle so I would embed digital literacy a lot more like understanding some it sounds almost basic right the sources like how are you thinking strategically and quite frankly as educator um I was focused on the more kind of the traditional reading and writing um arithmetics so I Ed digital literacy more and push that piece um in in terms of what I would do differently what I would continue to do and what I would urge like I think all of us to do in the context of education is one is um embedded intentionally into the classroom and for me to know it because even when I hear AI now as someone that sometimes I get irky when I hear like buzzword I almost think sometimes people just throw out Ai and whether they mean AI or not and what aspect you're talking about AI right are you talking about a tool are you talking about you know a specific system are you talking about a specific policy so as a teacher as a educator or former educator and I'm I'm at the Department of Labor working on Workforce and um you know as we work on you know report to the president around his executive order that I could touch up on in a little bit later is that being more intentional about understanding what the heck we're talking about when we say Ai and I would particularly focus on four areas and I would Infuse that into my classroom to me the four things area that comes to mind in the landscape both globally and nationally on AI and I'd love to kind of have the panel either push back on it add to it subtract from it is one is technical like safety like what do like the technical safety like what are how are we talking about in terms of um the tool is it safe what are unintended consequences number two I would say ethics and so there are really strategic decisions that whether it's government or the private sector or Civil Society they we'll have to make about the deployment of use of a technology or AI that I think sometimes kind of like social media we think about after the fact so the second thing is the um the um ethics the third thing I would say um is um uh is around privacy and so privacy actually in in many ways I remember when I was at the state um when I was at the state level working at the department of uh um labor in a governor's office at the state level I remember legislators and even parents and families like I don't want you to use my kids day you know I I I want to protect the data and one of the basic things I think we mostly know here is that when we talk about AI as if this is mythical thing it's really a bunch of data and then once you add in some like very Nuance where they deep learning large language models that's like building on to it and then the fourth and final thing I'll just say is um around bias like being very very intentional about that and it I really like Dean your point around like what are the problems we're trying to solve that you mentioned what are the questions we're trying to get answers to and the reason why I mentioned the bias and then I'll stop is because whether we're talking about as a educator or we're talking about in the policy landscape public and private sector we're using these tools to make very like to make decisions that in some ways we almost think like it's automatic like we think like they're smarter than us in a way and they're pulling from information that is already biased so whether we're talking about um you know um the you know if you do a simple search on you know whatever your favorite Search tool is and it says and there was one example like show me like a family or show me uh an employed person it will pull up a white family or white person um and I could tell you run through a whole ton of other examples both on the positive or the negative side but I think like the bias part is really important because it creates and it could create riffs in like both in political context and in communities that exagerate exacerbate tensions that's already present in a way that actually doesn't get us to answer the question that we were trying to solve to begin with did you want to add on that Paul because you were like nodding yeah I mean I think those are great points I think um you know safety ethics privacy and bias these are critical areas that need to be investigated in um education also in other contexts but they have a particular value in education um one one thing that makes the education context unique um in the United States definitely but also in other countries is that there is kind of a regulatory um Foundation already that addresses some of the privacy concerns uh in the United States we have a federal law it's called furpa that governs student privacy we have 128 state laws that add additional requirements on top of furpa so when we're talking about AI in the context of Education um part of the equation is already solved for U now maybe it could be better maybe it could be more aligned it's quite messy for for companies to deal with but we have that in place which is very different than the Consumer context and Dan you're you've been interested in some of the developments on the consumer contexts in the United States um and so um because AI is fundamentally based on the data um we can kind of build on that and really focus on the applications and things of that nature um but there are so many things that make education such a unique use use case that it requires a lot more attention and I think this is one of the reasons why we haven't seen a lot of governments come out with hard and fast rules uh for using Ai and education Japan a couple of months ago came out with um new guidelines that said you have to understand how AI Works before you use it in the classroom uh but that's kind of an anomaly um in the United States we've received some excellent guidance so far from the Department of Education um a really amazing report that came out last year I think it was in July which I encourage everybody to look at um we also are seeing some real collaborative work among Civil Society government and Industry um one effort I'd like to highlight is the Ed saafe AI Alliance uh which we are part of um which has a framework that is grounded in safety um accountability transparency fairness and efficacy uh and the whole goal there is to really pull together um experts pull together industry talk with governments domestically and internationally and try to uh raise the bar on AI in the classroom and try to um move towards real concrete guidelines to make sure we're doing this in the right way and I think that those sorts of projects are really essential going forward yeah can I um I I want to amplify too what what Manny had has to say on uh on equity and bias which is sort of extremely important um particularly for the World Bank too because we're looking at you know 180 Nations and I think even even the language that we use to create these these models is is very important and uh there kind of innate bias because a lot of that those those models are be being constructed here so if you can kind of picture yourself as a kid in in Bangladesh using this and then what comes up is the you know is the it's the you know picture that you mentioned um that you know kids's going to be rightfully wary of of that and then also too and and picking up what you had to say um in uh I think particularly in the Middle East in the UAE and and in uh Saudi Arabia you've made a like incredible effort on how to use AI properly in in education I think in a lot of ways more more so than here in other um kind of Western countries so I you know my guess is there's going to be a lot to be learned uh from that so thank you uh on that note as well as building on what my colleagues were mentioning there's a part on the data security and the data safety but there's the educational element that might be looked at at as biased but there's also the cultural and value angle to it and the language as well like for us in the UAE as we're building our tools it's not just building the applications it's also building the infrastructure it's building the llms that understand our that understand our language and be it Modern Standard Arabic or the dialect that the students the students use at home it's being able to to develop a tool that that they can speak with and they can trust as well as students and as Guardians as parents and teachers and that goes back to the critical thinking I would call of of a skill more so than just the digital literacy digital literacy is very important but also making sure that you have the foundations right and then making sure that your students as well have that skill to be able to judge whether the information they're getting or the answers they're getting is is correct or not and working with the educators to assess because it's going to be a long process We're not gonna get it in one go right we have we have and we can't wait until we have the perfect tools to launch them to the to the students so it it has to be a balance of of both sides just briefly on that um I think that's a real positive model for the rest of the world um building on what Dean and and Manny said uh if we're building AI tools on data that's not representative of the local local populations and of the cultures and values of the communities in which the AI tools will be used especially in education they're going to be of limited use um we've seen this come up in the healthcare context a number of times but education's even more personalized in a way when we're talking about actually using um AI to improve educational outcomes as as opposed to you know perhaps administrative uh purposes in the school um so so it's a real concern um everyone knows about the digital divide but um there's a real concern that the AI divide is going to be um much greater uh given the cost required uh to develop the kind of data sets that you need um to build those educational applications on top of them um and and I think that's something domestically but also from an international lens um a real issue and and to that point um Paul and something else RJ mentioned is one of the ways and this is like the the plug in in terms of the work that we do at the Department of Labor and something that I think we we don't want to see um and Paul you mentioned around the Divide I think one of the other pieces is so while there's all those other concerns I do want to say and this goes back even to your original question around like what we like what I would do differently or what we should be doing is making sure that students across the board actually have the exposure because I often think about in the Workforce Development context so we oversee the public Workforce system um at the Department of Labor employment and training Administration and we do a variety of kind of job training skills um tools and you know to um both on a policy and investment side but I often go back to Something in an experience um uh that something I heard from a student once and when we're trying to talk about how to expose them to make sure that they're aware of all these kind of new and emerging sectors a student said how can we get experience if every job we apply to requires experience and so I think about that in the context of what Paul what you mentioned about the divide and um RJ which you mentioned in terms of the um uh like how we to to ensure that people have the access to it because what we don't want to happen is that in 10 years we're back at this same session or maybe a different session I'll probably have you know a little bit more gray hair or or so where we're talking about like all these students that either had there's these new Industries or careers that were developed and some group of students have access to that because they were exposed to it while others don't for whatever reason it could be a resource conversation it could be you know folks you know you know policy landscape that didn't create the right ecosystems ensure that the um the individuals or the communities had access to those types of experience and exposure to these tools that will inherently shape the labor market and the in communities in a way that we want to make sure that it's you know intentional inclusive so so I have like another question when we are building guidelines and regulations how do we make sure that like uh the people who are involved in this process that they we have like them equally um involved so is this something you take care of especially when it comes to the executive order so who's actually on the table uh doing this I think this is one uh important question especially since we know that the data in all the large language models that's not equally um distributed so is there anything you can do to make sure to have all the voices on the table absolutely so I'll just mention really quickly love for my um colleagues that Chim in too and in terms of their perspective um but a few months back uh President Biden signed an executive order on AI and in that executive order um he task um uh agencies to do a variety of key things I'll talk about from the Department of Labor's perspective and of course we were closely with the Department of Education as well but the two kind of big picture things that we're really working on and we have this report due to the present in a couple weeks um I'm still finalizing um with the team but one is around principles and best practices and so what are kind of big picture key kind of you know themes or principles that whether it's developers employers instit governments um should leverage and then two working on a report around um impact of displaced workers and how we think strategically about supporting um um and how can the ecosystem really support workers that are displaced for a variety of reasons in particular specifically in this context for AI um that's been impacted by technology so going back to your question though about what could we do I think it's like the number one thing it sounds almost Elementary but it's like one making sure that the people that are most impacted or that could be most impacted they're at the table one of the things that we're doing and that we've been doing is that our team has been doing is whether we're talking to the developers and making sure that we're hearing from developers and employers but also making sure that we're hearing from workers and individuals that you know that um are most likely to be impacted I often think about from the policy context so when I left the classroom about um about 12 or so years ago and started Focus deeply on policy I often think about when I'm having a policy conversation I look around a table like one what is a problem we're trying to solve to um I just look around who who's most impacted and are they represented in that room or in that conversation and if they're not represented in that conversation I go out and try to seek their input and I think we could find ways to institutionalize that that kind of work and so at least at the Department of Labor we're literally just hearing from workers hearing from employers hearing from developers to actually both better educate oursel but also just to make sure that we're actually hearing the different perspective of those that most impact it can I just add on that um so following the executive order actually um I think it was earlier this month or or last month uh the Office of Management and budget which is part of the White House U issued a uh guidelines for the federal government's use of AI and it builds in some processes to gather input from potentially affected communities and be very representative for any type of safety or rights impact AI that the government ends up adopting and this is separate from what Manny was talking about but it's part of that bigger effort that the US government is promoting uh I think um you know my sense is is there's also an interest in Industry generally and I can't speak for every company uh to try to do things uh better and safer and more responsible um and uh a number of major AI companies agreed to voluntary commitments that the White House organized there's a G7 International Code of Conduct that many countries have agreed to and the education space we convened a bunch of edtech companies to issue a set of principles that we hope will guide the industry and again raise the bar on um how to make sure that AI is developed um safely transparently um with respect for Equity inclusion and and and so forth um so um you know I do want to say that in just one one point that in this space uh we can't just rely on government it really needs to be a much broader effort and um you know industry has an important role Civil Society has an important role um and I think we all need to work together and it's got to be more about just um guard rails guard rails are absolutely essential uh but if we want to unleash the potential that people over the past few days have been talking about endlessly um uh for AI and education we actually have to invest in infrastructure um but do it not willy-nilly make sure that it is safe and responsible but we really do need that that commitment yeah uh totally second that so you mentioned attex so to you rad and Dean um because you're working with attex like in in UAE and um maybe you are still advising atex or startups if you would um advise atex startups about AI on a global scale what would you tell them how should they work what should they do maybe you want to start rat uh I have a a follow-up note on the global landscape the UAE specifically around the reactions on introducing tools on and uh AI policies uh if I recall correctly it was like February 2023 when we announced the the ministry's uh AI tutor and the that we will Embrace AI in education we got two reactions from the teachers one is that what does that mean for us are you going to replace teachers uh is this like a a a robot that was going to go to the classroom and teach instead of us and uh two okay I want to embrace AI in my class classroom what do I need to do and uh this taught us very early on that we can do this on our own that's why we work very very closely with Educators when it comes to drafting the policies and guidelines and we did release our uh initial uh policy and guidelines for teachers on how to to use AI in the classroom with teachers with students um that's one and two is also when we develop our tools we need to to have them as well part of that process for example with the AR tutor that we're developing right now we have teachers on boarded as the uh Master Teachers that will help us develop the tool in a way and train the the tool in a way that that reflects what the best teachers in the UAE would do and to complement their efforts as well it's going to happen outside the classroom not inside the classroom and and replace teachers it's providing access to students who lack that we already have a global uh shortage when it comes to teachers so Technologies are here to support us in in for these challenges and not just creating challenges within within the education space or any other sector now going back to your question on my advice to e tech startups we've talked a lot about uh data security and uh safety uh throughout the the conversation today um for example in the UAE we have very strict uh data sovereignty laws and uh data security laws as well and if I would have any advice to a tech startups private sector and uh non-governmental entities as well to find the right partners because you can't find all the answers by yourselves for for example us as a Ministry of Education we don't come up with all the rules and regulations when it comes to data privacy and data security we work with the national uh cyber security Council and then we come in with the education specific laws and regulations uh so my advice would be to find the right partners that understand the markets that you want to get into and work with even with the Govern understand the markets that you want to get into and what their laws and regulations are thank you anything to add Dean yeah I I first I I don't want to leave you with the impression that we're not sort of educating um government uh officials and in our case that's who we work with um but we we developed uh several years ago a what we call a policy Academy so twice a year we bring in policy makers and and sort of their um the the project managers within the World Bank who work on um projects in those countries um that have a technology element and we go through basically it's a project-based learning course and um kind of bring them to a better uh understanding of what's um what's going on but if I mean if I my advice on to an edtech company would be kind as I mentioned before I mean there's tremendous demand and that's just going to keep growing because there are laggards and you know 10 years later they're they'll only kind of start doing this and I it runs the Gambit um as long as you kind of know what you're doing with respect to to to AI um but also very important to as R.J said too on the you know on the market to understand what markets you're looking at cuz you could you could waste an awful lot of time so you you know you want to look at the folks who are who are leading and uh who has started to implement some of this and who have the infrastructuring capacity to do it thank you so much so I cannot believe it but we are already at the end of the panel so I would like to ask you to just give like either one word or one tweet just what do you hope for AI in the future what is your hope uh maybe we'll just go here you start Paul I think we need to work across borders um I think we need more uh cohesion globally to increase the capacity to actually take advantage of what AI can do in the education space um I would iterate on the opportunity on ech and AI in in education and I know there is a talent shortage when it comes to AI so I would prioritize what solving for uh AI in education um and graduating the talent that we need be it the basic skills that we need for everyone to survive in the future Workforce or thech expert the AI experts that we need in the education space and other areas as well I think from a Workforce perspective I I'll mention another quick quote that I've heard from this one was from an employer um I remember he said we hire for hard skills we fire for soft skills and so in the conversation about AI let's everything that Paul Dean and RJ talked about in terms of Partnerships in terms of like all the components is also really critical and making sure that there's exposure but on the other hand let's also not forget in the context of when we're talking about working with children working with families to make sure that we're complimentary in the conversation um when we're talking about AI so that we're always combining both so that they um have the most opportunities to be successful yeah I think in in the country countes that we uh we serve um it's clear to me at least that we're just not going to be able to build enough schools train enough teachers hire enough teachers and so um I I think the realization is getting there I mean if we could have we I hope we would have but we haven't um so I'm I'm actually optimistic about how we might use AI to kind of close that Gap but you know we have to be very intentional about it so thank you thank you very much to the four panelists a big round of applause [Music]

Channel: Global Silicon Valley

Views: 227

Keywords: arizona state university, asu+gsv summit, conference, edtech, edtech conference, education, global education, gsv ventures, skills summit, technology

Id: ciCayZJANu0

Length: 42min 10sec (2530 seconds)

Published: Sat Apr 20 2024