President the Tonga women in ICT Selu should be here, but he's not. She's not here due to other commitments. But thank you, Miller for taking up the role. So I like to call mela to please make yourself here to the podium and present your the Tongue and digital inclusion and capacity building. Thank you.
Applause.
Malo ele and greetings to you all. My name is melitavake, and I'm currently working for Tonga women in ICT, and thank you to PASOK for allowing me to be here as a Pasc fellow. I acknowledge also twi court members that are here in the room. Doctor, thank you, Nate. And I'm happy to present the work Tonga women in ICT does in Tonga around digital literacy and capacity building, inclusion and capacity building in the Pacific, particularly in Tonga, to ensure that all communities are encrypted to participate in the digital economy. Before I start on with my presentation, I would like to give you all a little brief about our town, women in ICT and how it started Tong. Women in ICT was as the place in 2018 It was started with a group five women's. This little group were five women's. Those five women's, they came up with an idea, how are they going to help Donga, especially for women, all of them. They're from different organization, and they can see how those organization underestimating women's. They said that all the works in this organization, only men can do it, but this women's, they came up together for a cup of tea, and that's the thing they were discussed. How are they going to change Tonga, especially for the women's, they came up with a little name. They said, Maybe we should name ourselves as a twit. So twit stands for Tong women in ICT. The reason why they named the little group as some women in ICT because all these women, he came from an IT field. I all
those
churches and stuff, it was taken from their own pocket. There was no budget in this time until they all became financially stable and they managed to register Tong women ICT in 2021, 2021 they were over members, 30 members. I can say, without these women's I cannot stand in front of you guys here today, and I can say that these women's, they are digital warriors. Why?
Because they encourage us women's to be more mindful. Not only that, but they encourage us to be smart in using space online, especially ICT I
Tonga women in ICT came up with a theme which is strengthening digital covenants and resilience in Tonga With a focus on these areas, first, digital inclusion, capacity building and innovations and entrepreneurship. Tong women and ICTs missions and visions. Their mission is to empowering women in Tonga through ICT in a range of exciting ways and fully participants in the ICT industry. The vision is for digitally inclusive resilience. Dong Lang, which means allowing women and girls to enter the ancestry without fear or discrimination or persecution all those allowing them to improve it that vision and mission are to accomplish our objectives, which are to promote digital literacy, to build capacity for women in ICT and to foster innovation entrepreneurship. So exertation is to foster digital inclusion and building capacity. I city across various age groups. In Tonga, we engage as people of all ages through workshop, trainings and training of the trainer programs to ensure digital skills and accessible students, not only the students, but the young students, youth and also people that are older in the community. Through adopting a Maori stakeholder approach, we aim to enable innovation and entrepreneurship, which are currently limited in Tonga. And through this effort, we hope to inspire locally driving technical ICT and internet solutions that strengthen digital governance and resilience in the Pacific Islands. Thank
is for the key initiatives on digital inclusion programs, as I mentioned about one of our key initiatives is engaging all ages through workshop trainings, awareness and stuff. Secondly, is making teacher inclusively fun for interactive like and enjoyable learning experiences to maintain interest and promote sustainable engagement. Firstly, in digital literacy,
also ton of women in ICT, we do competition, a lot of competition. There was one competition that was on last month, which is an essay competition, the topic for it was bridging divide and empowering youth for digital skills and innovation and digital entrepreneurship.
For our capacity building efforts,
we do a lot of workshop, as I mentioned about, a lot of trainings, a lot of trainings that that's what we do to strengthen our capacity building in Dong, we do a lot of capacity building activities to enhance our digital skills across the community. Through that, we teach in trainings, basic trainings, on Word Excel, and PowerPoint,
not only that, but our promoting innovation is
we conduct a multi stakeholder approach through collaborating with various stakeholders to create a supportive environment for innovation and entrepreneurship in Tonga for locally driven solutions, we encourage the development of Tonga technical ICT and internet solution, fostering a sense of ownership and capability within the community.
These are images from some of the events that we conducted this year. We invited encourage you all to join us in our mission to empower the Pacific Islands through digital inclusion and capacity building. Our calls and impacts through our initiatives, Twix aim to enhance digital literacy skills across all groups in Tongatapu and also the outer island, not only there, but the other island in togatapu, which is wabao habayi, Iowa, and the two newa, we enhance digital literacy and skills across all ages. Group in Tonga, foster a culture of innovation and entrepreneurship by providing a space, we also strengthen the resilience of digital governance, working together with other stakeholders, the challenges, the challenges that we are usually face, core funding, we always lack In funding, skills development. We always wanted to build Tonga women in ICT skills development all the time. And the last we are short staff. There's only two full time staff, everyone else their part time. So we are short staffed, and the need for capacity building of our team is also important. And for our future plans, we wanted to expanding our programs to the outer islands, wabao, habayi, Iowa and the two news we wanted to do and enhance our partnership Within the togatapu and also across the Pacific Island and continuous support for innovation and resilience.
Malo apito I
thank you very much, mele for that fruitful presentation. Any questions for mele, but we can also ask mele questions after the next speaker, because we run out of time. If you have questions for melee now you can ask after Andrew from China, so I'd like to call On Nancy our next speaker for her presentation. Please. You
um Kiara, good bloody. My name is Nancy. I'm from Papua, New Guinea. I'm a fellow fellow for the Asia Pacific Regional Internet governance, and also for the Pacific IGF.
I currently work with Department of Information and Communication and Technology in Port New Guinea, the under the ICT ministry, we have the regulator and policy. So I come under the policy. Before I came up here, I was just talking to my fellow from png Russell that you know, after the usually in PNG after the every workshop or forum that we held after the lunch time, the food usually kills the people, so it's the boring part of the session. So I hope I don't make it boring. I know I have a lot of slide, but they're just bullet points. So my presentation is going to be about data governance, and that's data governance report from public sector in Papua, New Guinea. So currently we don't have any act or we also don't have any data, governance, national policy. You
okay. The purpose of the assessment report was to evaluate the current data governance practices across the all of government. That's all of government approach to identify strength weakness and provide accountable recommendations to enhance best practices of data management, ensure data quality and comply with regular requirements. So this assessment aims to support strategic decision making, improve data integrity and foster a culture of effective data stewardship across all of government, because in PNG, currently, we don't have any national app to protect our data or even govern our data, And also, we don't have any national policy in place. So currently, the department that I worked with, we have just got our first data governance and data protection policy that was approved, and we in we're In the states of drafting instructions for a act. You
the overview of the data governance framework from in the public sector that we're driving is so the department coordinate from a national level. And so what is data governance, and what is data governance looks like in any organization or any person. Point of view, obviously, program that focus on privacy compliance, security and how it looks different from one that may exist to support data warehouse and business intelligence, and a program concentrating on architecture or integration may involve different participant, then we depth one whose goals involve data, quality, whatever you focus your effort on, I think we need to define for our participant, a exactly what we at our national level in PNG. Mean by data governance. So we decided to choose our words carefully and must resonate, resonate back to PNG, culture and environment, if this is to be accepted, knowing very well that data governance is a new concept in the public sector and a new agenda also, although the practice has been around, but not Much has been researched and talked about data governance, i So, so it means different things to different people, organization inside the public sector, back in Papua, New Guinea. So we, from our hand, we think that it is important to present a definition that implies to what we are trying to accomplish. So basically, we said data governance is defined through our structure at the national level, within PNG, and also to what we want to achieve across the public sector or all of government approach. So it's a set of principle policies, process aimed and ensuring data is measured, sorry, managed, effectively, securely and responsibly throughout its life cycle. I
so one of the one of the effort that the department did was we, we, we've been holding workshops throughout the year that data governance division that I worked with, I worked as an assistant data governance manager, and my role is to lead data governance all across government, right? Government Sector. So we have all held couple of workshop throughout the four years, but most of the workshop, it was more about us telling the public sector what we want them to do and what we are doing, and we didn't listen about what they wanted to do and see how they're managing their data. So we headed the workshop, a national workshop on the fourth and fifth of June in portmore. That's the total number of participants, because it's a government workshop. We have about 55 total agencies, all government department, government department and national agencies, with about 101 participant during the workshop we we try to assess by letting giving the stakeholders changes to speak, so we get to define and understand what they do In regards to data governance within their own department. I
The key components of data, governance, assessment during the workshop that we came to see was we've categorized it under three categories. The first one is institutional framework, we and then oversight and accountability, legal and regulatory framework, the Assessment Overview of the report that came out from that we also have a sorry, we also have a online survey, but the survey with the workshop, we did a national data governance baseline report that is currently still in drafting, so I'm only presenting for the workshop that we did. The assessment objective of the report was to promote responsible data management across public sector and then stakeholder collaboration understand the current Data Governance landscape across the private public sector. The
method methodology of the assessment was actually the workshop, because that was the only place, like a platform, that we could understand what other public sectors were doing in regard to data governance across all of government approach. So that's the key findings. It was very interesting doing, while we were watching, just sitting at the back and seeing all the government department doing their presentations on how they manage their data,
30% of the agencies have some form of data protect data governance practices that includes data collection methods, storage infrastructure, data sharing, data sharing, cost and optimization, data accessibility and availability under oversight and accountability, 50% of the stakeholders or the agencies have data protection measures include. That includes access control, which is the main form regular security assessment, and 20% is compliance. So you can see that other 80% are unsure of regulatory requirement or national standards and best practices. We also have data retention and archiving, so most agencies like formal like probably 90% of agencies like formal guidelines under the legal and regulatory framework. 60% use data security measures. Backup and disaster recovery was very little. We couldn't even hear a lot of them say anything about backup and disaster recovery, data sharing, absence of data sharing guidelines and policies was mentioned nothing.
So we have found out that you can see that most, most of the teams was just technical. Nothing much about data governance. So this is the maturity model that we have to measure across. It's still in we're still developing it, and it's taken cue from the Kenya data governance module. I
I tell it. It's tailored for the public sector in PNG, not the private only public sector. It should be simple and editable to read story adaptable to the current environment. It's quite the level stops at five, but I'm just putting this as an example still in progress, for contextualizing and we need to include our national pillars, like the institutional framework, oversight and accountability, technical and regulatory framework can have some index to measure it. The majority model is also taken insight from the PNG Digital Government Act. 2022, digital government plan, 2023, to 2027 the medium term development plan and the national data governance and data protection policy, which was just approved. So these are the recommendation that we have come up with from the workshop that was held still under the three categories of institutional framework oversight, accountability and legal and regulatory framework. So under it, we have, under the institutional framework, we have standard, some framework. These are the recommendation from us, Department of ICT, to all the stakeholders that have participated in presented we've seen that these are the things that are missing within their data governance practices, so that standards and framework develop, develop and implement data governance and protection policy within their own sectors, and the Oversight and Accountability increase the frequency of data governance, review and audit, training and support is Also one of the bigger things that we've like developed, data protection legislation. Most departments doesn't have any internal policies in regard to data governance or any legislation like the country, we don't have any yet. And then data protection measures, and then compliance with regulate, regulatory requirement. The next step from after the workshop, after measuring all the doing the assessment we have from the department, we have decided the next step was to establish a data governance working group, coordinate and implement at the national level. Thank you. I also have last something to say so lastly from a Asia Pacific IGF fellow and the Pacific IGF fellow are not from my employer as a government employee. In IGF, we promote multi stakeholders. I think in my country, we need more awareness on IGF for private, private sectors and civil society to know the importance of this in the government policy, because my department does the national policies for every high city, and we have seen lack of private sector business organizations coming in to take part for the policy. And when the policy is abolished and becomes policy, or even the act, they get left out, and that's one of the bigger things, and then they are not included in them. That's where so many problems can help. Thank you.
Thank you very much, Nancy for that sharing of insights about data covenants, and it's very important these times with internet covenants that sometimes we focus so much on the technical side. But when it comes to the layer seven, with the applications and data transferring, data sharing, that's where the main problem comes in. So but it's good to see that png started their assessment, data, governance. I think some countries in the Pacific, they did that as well. But it's good that you consider the private sector, because with this forum, we promote the multi stakeholderism. That was my question going to ask, What about the private sector? But it's good that you address it before you finish. So the next speaker, it's Andrew Andrews from China, and he is leading the China IGF, uh, we're giving an opportunity now because we don't have another our another speaker is not here after him, then we can open the floor to some questions. I have some questions for me as well. Please populate your questions for this session. We have so much time, so this is why we here, so we can share our insights. You know your strengths and other people's weaknesses. Maybe that can match and solve the problem. So please don't be shy. Over to Andrew.
Here is the controller.
All right,
let me try this a little bit to see if it work.
All right, all right.
Hi everybody, good afternoon. First of all, self introduction. My name is actually my Chinese. Name is guotong. Andrew is my English name given by my actually by my middle school teacher, but I haven't used it very much. All of you can call me Feng Feng. So, so I'm from the China IGF. I'm but the work of China IGF is my part time job, my primary position is from the China Academy of ICT, which is a research branch under the government the Ministry of Industry and Information Technology of China. So I'm also serve as the China's representative, advisory, representative of Governmental Advisory convened committee in the ICANN so I'm currently the participating ICANN meetings regularly. So I'm glad to be here to attend this Pacific IGF. It's a very good opportunity to for me to interact with you and to learn from you and find the community of Pacific IGF is so Oh, it's full of energy, full of motivation and with solidarity. So I think I can take this opportunity to most of important is to learn from you and share with you some activities and experience of the China IGF, so I'll give you the presentation of the overview of China IGF, I Yeah. So first of all, who we are. So we are. We are the China IGF. So many of you may know that the UN IGF, the encouraged regional and national IGF to be established within country, within territory or within a region. So we are IGF. We are a bottom up and open platform for exchanges and consultations on the internet governance related issues. And our mission is to promote the Chinese Internet communities in engagement into the global Internet governance process. And the we have a secretariat. The secretariat is taken by the Internet Society of China. The Internet Society of China. It's like a carrier for the China IGF. So we established the China IGF in 2020, in May by. I'm one of the, actually, I'm one of the co founder of the China IGF, and we have Ms peiwei, you can see, as you can see in this slide, she's another co founder of the China IGF, and Mr. Jungwei, I will show you in the following slide. You so what are we doing? Our tasks are first of all to promote multi stakeholder based participation and collaboration on internet governance issues and on internet policy side, and we also conduct international exchanges and cooperation. We participate to the global Internet governance processes, mainly to the UN and like the process here the Pacific IGF which is a great, great one among all the local and regional IGF. And number three, we we also we focus on local priority Internet governance issues. Like domestically, we are facing a lot of regulation issue and coordination issue with with regard to the Internet governance, and we also deliver our voices and deliver the voices of Chinese community to the global Internet community. We participate the UN IGF and its processes so the leadership of the China IGF now we have a committee. We have a committee, China IGF committee, and also we have a China IGF mag, which is a China IGF multi stakeholder advisory group, which is very similar to the UN IGFs Mac. So Mr. Now, Mr. Gao Xinmin, who is the chair of China IGF Mac, and Mr. Huang Chongqing, who is the former president of the sensor, who is now the chair of the IGF committee, and who is now also a vice president of Internet Society of China and MS Zhang Xiao he. She is from sydnik. She is the Deputy Director General of sinnik, and also serves as the vice chair of China IGF and and she is currently a member of the UN IGF mag. The next is me and zhongrai. Mr. Zhongri is the Secretary, Secretary General of the China IGF Secretariat, and mainly he lead the team, also including me to coordinate daily issues of the China IGF. And me, personally, I do some support work to the previous leaders and also amused. But my main main work, daily work, is either China ICT, China Academy of ICT. And I'll give you some highlights of China IGF. And we, actually, we organize a lot of workshops related to internet governance issues, whether it is in Beijing or other part of China, and we perhaps we organize workshop workshops in a big internet conference or a local internet conference within China, and workshop, seems, I believe, is including child online protection and digital inclusion, data governance and information accessibility. And we also organize workshops within the UN IGF during the UN IGF meetings week. And we also contribute a lot to the international fora like un IGF and other related internet meetings. We have, as I told you, that we have multi stakeholder advisory group, which includes experts from different sector, like from university and from the industry, and also from, from some think tank and those experts are about 35 persons, persons. The also participate a lot of international and domestic meetings, and they represent China IGF to to speak, to contribute, and also we conduct as many as we can, to to outreach to people, to have engagement to the community, different community, like we organize some event in UNF and also, perhaps in the WIC in China, world internet conference in China. And also we have yearly based internet conference of China and other local industry. Internet industry conference, we may acquire a boost to show to the to the audience and participants, what is what is all about? What is about the China IGF? So this slide shows you the how we participate in the UN IGF related events, like we participated the IGF national and regional IGF meetings. We joined the talks and to share our activities to the
to other participants. And also we, we have Miss John Xiao, who is currently the member of IGF un, IGF Mac, and we previously, actually, we have several others, people from government or from from industry or from academic society, the previously they were the members of the UNH mag. So the join the Join the conversation going, join the consultation of the Mac a lot to set up the agenda of the annual un HGF, and also the logistics of how to organize the UNH workshops, the rules to The procedures to for for all of the multi stakeholder to apply for the opportunity of organizing workshops in ungf. And also we, we join the meetings of IGF, national and regional IGF initiatives, and we speak, as you can see Mr. Mr. Speaking here. And also, I mean, this year's is a quite a hot process and quite important process that that is the GDC global digital compact, which is proposed by the UN Secretary General, I believe, or un United Nations, is about the initiative about the how to govern, govern, the internet governance, or digital governance. So last year and this year, we talked about this process and the issues related to the papers and the GDC a lot, and we we also share our views, different views among the members of the China IGF, and we produce our input to the global digital compact, which is, I think, is effort from the China Jeff to the GDC process. And we are continuing this, this effort, this year, maybe, I think perhaps this CDC will be, will be held this year, perhaps this year, August. And also we try to produce some input to the IGF, p9, I say Pinay, which is a policy network on the artificial intelligence. Pinay, we have a leading expert who is the Associate professor from the China Normal University, Ms Cheng, who is leading this effort. And actually he is one of the, one of the leading persons in this, in this effort, and also, I think artificial intelligence is very important to attract all of the attentions from industry and also academic world, and also from the government, government perspective. So we, you know, we have, now, we have a lot of internet artificial intelligence startups and big companies operating AI services in China. And we also want to contribute to the global in an artificial intelligence conversation. And also we we share this international process back to Chinese community to let them know what is happening, what is going on in the in the UN and in the global processes. And we also produce some, perhaps input to the AI advisor body, which is a body related to the United Nations. We have our community effort to input to the this advisory body, like as a title, show you that China I have input to the governing AI for humanity, which is maybe one of the topics. We have other inputs to other topics. And this shows you the we organize international seminar. One of this is the global initiative on data security. It was, I think, organized two years ago. It was in the national seminar to on the subject of seizing digital opportunities for cooperation and development. We organized that and and we also have other, perhaps other international seminar to organize in the past four years. And we every year, like Pacific IGF, like you, we we host, hosted, we organize the Agile annual chinagf meeting since 2020 but as you may know, that 2020 is a year of the beginning year of the covid 19, so we organized our meetings online. So we have also invited un Deputy Secretary General and peoples from around the world and locally in China and our own China IGF Mac members to speak to to share views on a variety of topics related to internet governance and also digital cooperation. And last year, we began the effort of the physical meeting like this, in the in the in the in the in this right and below that, this picture shows you we have our physical meeting last year. So to calculate the number we have in the past about around four years, we have organized a lot of events. We have salons, some some informal exchange of views, occasion like 30 plus. And we have, we have been taking some training for more than 1000 and we organize a lot of workshops online and in person. And we connect connected to other organization, other community, or other other society, which is 10,000 plus. So we, when some awards like this, we awarded as outstanding case of jointly building a community with shared future in cyberspace in 2023 in the world internet conference. Ms Zhang Xiao was representing China Jeff to show, to show the case of China IGF, which is big, I think, big honor for the whole China Jeff's member and the Secretariat and that's all about. I want to give you about the China IGF activity. The last one is, is the, perhaps the sign, the identifier of these China AGF, this is the sign of Internet Governance Forum of of China, China. Jeff, so as you may see that we have a lot of Chinese culture elements in this actually, I was leading a small team to define this design. As you may see that there's a panda over there. So panda is kind of representing China, especially Sichuan in China. It's very cute animal. And as you may see, the English word China, which is conducted by the Chinese calligraphy. Chinese, we call it the Chinese pen, Chinese Maui, which is hair hairpin, Chinese hairpin. Chinese hairpin. The hair is whether perhaps is made from rabbit or sheep or even wolf, the Chinese, Chinese style of way of calligraphy, the Chinese stroke, that all the strokes are conducted by the Chinese mobi. So
actually, I asked my father to to conduct this, the China the stroke, because my father, who is really love to write chance calligraphy. And as you may see, the China, the in the middle I which is like a seal in China, like a seal, the in the middle, there are the two characters, which is dongguo, means China, and those characters are traditional Chinese. So I think this is, this is what I want to show you. And I really enjoyed this two days meeting. And hope to hope this perceived IGF, a whole success. Thank you very much.
What about another round of applause for Andrew, all the way from China to present the China IGF to us here in the backside. GF, so now we come to the question time. We have mele presented the digital inclusion and capacity building, and we have Nancy, if you can make your way up here so that they can see you, and they ask their questions, and for you guys to answer it, please. I only have one question for you. But then I leave it to the to the audience,
as you see that this session is about a digital transformation in the Pacific. And if you hear this morning, the journey that the pika sock and the bike back I need went through, it's really interesting to see that in in the Pacific, failure is part of life. You know, failing approach. It's part of life in anything we do. But we learned same as the digital transformation. You know, we fall, we stand up, we fall, we stand up, but it keeps moving. And that's why we are here, because we learned from the last 1020, 30 years. We learned from the best, and that's why we come and we still striving, not stopping, but we strive for our Pacific community. So hence we have our beautiful fellows. Here they are all both Baker GF fellows for this year. And I have a question for melee. It's about the capacity building and digital inclusion that Tonga or the tweets been doing in Tonga. Have you done any evaluations of the methods or the you know, the approaches that you've been doing in your capacity building events
so far? I
thank you for the question. I think for that question, I'll pass it on to Dr. Thank you nastin, so she can answer your question also. Thank you Now she's our board member.
Thank you, mele for dubbing me in as a board member of twit. I'm really proud to be here to represent the Twitter board. I think I'm the only board member here. So you were asking about the evaluation of the because the tweet has had a lot of activities right throughout the year itself, we going back in time for the results to be shown to us as board members. So yes, every time that they're having an activity, there is a feedback form in there, you know, for us to evaluate and have a look at those. I think mele was also pointing out the fact that there is the challenge for us, because with all of those activities that we do in Tonga itself, for improving digital literacy itself is funded by different people. So there is no single funder that we can rely on. So if there is going to be, say, for example, the camp for women in STEM we had to look for funding. We had to look for for a number of fund funders, you know, to sort of fund different activities that we are having. Even the last activity that we had, it was to do with improving the digital literacy of teachers. We also have to go out and ask the government for funding. Unfortunately, the government don't have the money, so we have to ask for donors in Tonga itself is Chinese, the Japanese, the Australian and the New Zealanders and also the US. Thank God. Some of the of our funding come from Australia and New Zealand, most of them, but because we are non for profit organization, and we are doing fantastic work in Tonga and trying to sort of change, you know, like the mindset of older people, as well as bringing in the young people who are digital, digital natives. The difficulty of trying to knock on the doors and getting money, money for funding. We just had what we call a steam Princess camp. And to run a camp, a residential camp itself, you know, is quite expensive for all of our people, you know, but we tend to have a bit of money, and we run those camp itself, you know, like so the evaluation, if you are interested, you know, we'll send it hopefully. You know, by the end of September, all our evaluation from the different projects are coming in together so that we can see our future plan. But if there is a anybody here who is a donor and Scott deep pockets of money, we are interested in having a lunch or dinner with you. Thank you.
Thank you very much. Any question, yeah, what can
we do to help you do 10 times more, but maybe 100 times more. What can we do? Sorry, I'm I must repeat the question so it's on record,
so I'm really impressed by what you do. It's fantastic. Work. Fantastic. Absolutely. My question is, what can we do to help you do 100 times more?
I think, as I said, you know that if the money is there, you know, we can, you know, maybe you know, like it's, it's not really a joke, because if you have money, you can plan, you can plan about five years ahead, or anything like that. But because you're only planning for three months of the time that you're funded for that particular project, you tend to be looking at short term instead of long, long term. So I think the money the other thing that we did find out is the shortage of staff, as mele was saying, is only two full time staff. And you know, I've just been notified that a princess of Tonga is also coming next Monday to visit the headquarters, you know, of a city. So we rent place in our church in Tonga because it's cheaper. The rent is cheaper. And I've you know that we basically have to ask for money for the funding of those area. We also have a training area that is limited to about 20. And because I also work at the University as the Vice Chancellor for the Tonga National University, which is a new entity in Tonga, normally when it's over 20, I just get them to come and use the university premises for that. And I think one of the thing, I think that came from there, is lack of staff. If you are running steam printers camp for 100 kids, you know, from about nine, nine to 15, or something like that. You know, it takes a lot of people to come in and volunteer at their time. So we go out there for people to come and just to volunteer, give them a bit of food in the evening, or something like that, you know. But, but we can't really pay them money. So money, I think, expertise, having the time to plan long term, instead of just, you know, within two or three months of those projects. So we are looking short term all the time, and there is nothing, you know. And when the the funding comes, it's only coming in to do a particular thing that you have to report to, you know. And it doesn't necessarily fit the the whole, you know, governance thing you know of having to try to improve the number of people that is IT savvy, you know, in in high school, so and as a melee was saying it ranges, you know, we started with teenage people, but Then the people in the community want one. So we need to also go out and asking for fun for community itself, you know. So if we're going to be covering the whole age group, then yeah, money is the thing.
Thank you. One question from Russell.
Thank you. My name is odeka. I'm a ccTLD, manager of.pg and also I'm a director of PNG, University of Technology. I have several question over this capacity building, but in this stage, not this area. So I have one simple question about the dx, this digital transformation. How many Pacific countries already applied? Apprised visa, tourist visa by online. Like, for example, I have a Japanese passport to apply New Zealand visa within two within, I think two to three minutes approved, and then I can come visit here. But many Pacific countries include our country PNG. It takes very long time. For example, my walking visa, still, I'm waiting three months over. Thank you. So how many other Pacific countries, if I have a PNG passport, want to travel to Tonga? How soon we can do it? I
That's a difficult question, because you're asking for, you're asking for a system that for our Pacific countries, that in terms of uploading, uploading your work visa, but I give it up to Andrew to assist with that.
Thank you, Russell. I think Andrew, I think we're getting there. I think I want to let you know that Vanuatu visas are now online. If you want to apply for a visa, you can get online. We just had that in a launch last month, so I think the Pacific is moving in that direction. Thank you.
Good question. Thank you very much for the opportunity and to our tongue in neighboring country, I have a question for you, please. Can you please state your name? Sorry, better from NUS Samoa, National University of Samoa, yes. And the question is, we've shared a wide range of activities and initiatives. Challenges you know that ICT Tongan always face, I may say, especially with the financial financial challenges, and also considering the the South Pacific region. And I may say that I can clearly relate to the challenges that ICT tongue also experiencing. The question is, what strategies have been implemented to secure sustainable funding?
What are some of the strategies that you have and
I think it's overall, it's really difficult to get sustainable funding without a track record. So that is exactly what we are doing now. Is having that track record that we can do things. We can make a change, we can make a difference. We can make the transformative digital movement that we need to have. And having that track record itself, then funders and also the government will start believing in our in our capacity to try to do this in toga. So I think the the like everything else with any project, you know, if I give you money for the project and you fail to deliver the outcomes and deliverables itself. You know, then I'm not going to be confident in giving you another pocket of money. So, you know, trying to build that up. And I think of the fact that we are doing a whole lot of those are basically to do with the demands from the public, rather than us having just to sit there and say, Oh, I think they need to have one. It's the demands from the public and especially the young people of Tonga. Now, you know, wanting to sort of have that, by the way, in order to access the internet in Tonga, they use their phone, and there are more women using the phone than than men. Apparently, men will have to find the money to buy the women their phone. So I think that's a good thing to do. And, yeah, trying to sort of get a strategy going is just basically what we as board members will have to look at, looking strategically in the future of how many people that we're going to be training you know, from one year to another to another, and making sure that we're going to consolidate, consolidate that, and asking you know based on those demands and the evidence that we have had, and having young people who coming in To witness you know, or or give a testimony of what they have learned, I think that would be something that will push the government, and also the funding bodies itself, you know, to sort of give us some more money and trying to establish a sustainable way of us having to do that. Thank you. Thank you.
You have two more questions for melee, and then we have questions for Nancy, because we have 10 more minutes before we go for afternoon break. So yeah, Leticia and then ashtosh.
This is not a question, but I think just to add on to what the response before about the challenges, I think one of the challenges speaking from sorry, my name is Leticia from Solomon Islands, and part of the women in ICT Solomons, one of the challenges I face, or we face at the Solomon Islands, is also writing up concept notes to give to donors. So it's not just skills on ICT training, but also the writing and skills reporting skills, as well as confidence and presentation to our public like this. Those are some of the skills that even within the ICT women group, we we need to train them on so that's one of the challenge we face as well. So yeah, looking for mentors who can help us in that space as well. Thank you.
Thanks. Ashtosh,
hi. My name is Ashutosh. I'm from VG. First of all, a big congratulations to the women in ICT in Tonga. I think it's a very good effort. Okay, my my question is, since, um, it is a very male dominated area, right? I mean, what challenges did the women of in ICT in Tonga face? I mean, like, do you have challenge with girls? You need to encourage them to come into it, or it's already there. The technology is there that is excites them to joining. And do you have like programs specifically for women, which encourages them to Okay? I mean, this is the path I want to take for it.
I didn't even know that I'm going to be here, and I feel a lot of questions. Yes, you will looking at our programs. Of course, the university offer formal programs for those people who are in form seven and going on to do a degree in ICT, so we offer a certificate, and also diploma and also a degree, together with another university in that's the formal education in terms of, you know, the gender, whether there is going to be a gender issue to do With the ICT. I think Tonga is very fortunate that we don't have those demarcation between ICT, although Andrew here from the government and E governments of Tonga may, may probably be having to answer to the fact that there is a lot more men in his department than females. But I think the ICT, the ICT the Tonga. Tonga women in ICT was basically formed because, as you as mele was saying, women in ICT having coffee and understanding that there is a need. There is a need in our local need for women to come together because they would have the same qualification as men. But men has been picked, you know, to have to work in ICT, you know, because of hardware issues or software, or perhaps they know more about the technology jargon or anything like that, to do with networking and cyberspace and security and all of those. But because it was a need to sort of making sure that our young women who are coming after us, you know, it's going to be technological savvy with their knowledge of ICT, and also to encourage them to take on the the you know, development ICT, whether it's development of apps or things like that, you know, but it was, you know, established to bring in the women. The men now are feeling really uncomfortable in Tonga because the women are now have advertised for all of these things, and then the young boys will say, How about us? Can't We? Can't we join the steam Princess camp. No, you can't. You're not, because you are not a princess. But you know, like it's that sort of momentum in the change itself. You know that is looking for for women. But I think Andrew, or to also answer about whether there is a demarcation in Tonga, about the gender, gender divide to do with ICT. Thank you, and I'm putting him on the spot too.
Thank you. My name is Andrew torman. I'm the director for digital transformation for government, and I'm just sitting here and listen to what collaboration is all about. And very interesting. One thing that I like to set the record straight, I think one of the biggest problem is that NGOs and organizations like women in ihcd, there's a little bit of disconnection between them and the government. And what I what I mean by that is that the government have a lot of projects going on, and some of the components of the projects may need to utilize the capacity of NGOs and organization like women in ICT, but because there's, there's not much information that we get from them, they Don't apply when we when we put up the bid online as part of the procurement, so they kind of lose out, but they're only there are some they did, like, I think Andrew knows about that, that they come straight to women in ICD. But I think if we, if we can establish that connection, it will be something that the government can also help from our side. As far as the concern about the women in the industry. The problem is women kind of lose interest when they go out, more technical, like programming, developers and stuff like that. They're very good in high school, and when they go up with the high school, when they go up to university level, they're gonna lose interest. They probably looking at other areas that making more money, business and stuff like that. When we when we advertise our jobs, we wanted to have like developers. We wanted to have, you know, women that get into cybersecurity, but unfortunately, only a few qualify in that specific area. We have scout in. I don't know about USP and other universities, but there is an evidence that a lot of people are not going into technical fields, and we kind of think because they have issue with math, a lot of them are going into arts programs.
So it
is an issue that we notify from, you know, notice from, from the government side, we were doing our best try to attract more people. What we have noticed at the school level, there's a lot of female graduates. They're more than the male graduates from high school. But then the different interests when they go to college, when they go to university, then there's a team occasion there, they pick other attractive programs to pursue, but we are welcome if there are female who wants to join The team. We've been looking for people. We're not discriminate. Them because of their sex. So thank you.
Thank you, Andrew. We have one more question, because we run out of time.
Maureen, thank you very much, Sarah and Andre for giving me this opportunity. My name is Maureen Hilliard. I'm actually here sort of like representing the Board of Trustees of the Internet Society. And I've just taken on board the gentleman who just spoke before me. I work in the area of project development in the Cook Islands, and there are heaps of projects going on with across the Pacific. One of the things I think is really important is if you find out there's a project available, is to get in there and find out where the gender component is of that project. Every project is supposed to have some kind of gender component, and sometimes that isn't actually emphasized enough, and it's great that it was raised today. So again, I mean, I think this is one part of putting, you know, ensuring that women are involved in the development of your country and through the projects that are actually taking place. But what I really wanted to raise was that pick ISOC, as a as an Internet Society organization, is entitled to apply for funding through the Pacific, through the foundation, you know, through the ISOC Foundation, putting, you know, I mean seeing the rising, sort of like trend of women in ICT, which has been long term, you know, if you put forward a proposal that reached across the Pacific, and, you know, really encouraged the and I know there are women in ICT groups in a lot of the different islands, anyway, but I think that a major project across that pikaisoc actually ran would be beneficial to all the groups of women within the Pacific so I just want to put that forward. Thanks.
So thank you, Andrew from pikaisoc and Pacifica JIFF, I just wanted to maybe say a few words and what was discussed here. So we have had a lot of these groupings around the countries. We have, excuse me, in Vanuatu, we have this and we have the smart sisters program initiative. You can find that on the Facebook page. We have Tonga, we have Solomon and Samoa, but in response to what Maureen said about pikai sock, you know, finding opportunities, we have done that in the last, I mean, since last year and this year, and we have got the support of the Cade and European Union that included funding, partial funding, for Pacific HGF women in ISD in Tonga and Solomon Islands. So this is the first phase that, and they're going to run for two years, 25 and 26 so there is something that pikesoc has done that is becoming coming to fruition, and I believe that could be a lesson learned for others who are interested to support such initiatives. Thank you.
Thank you, Andrew, for that. Any questions for the data covenants framework before we break for afternoon tea? Otherwise, you have plenty of time to ask this beautiful woman with their presentation they presented. Oh, do you have a question? Okay,
thank you. Hi. My name is cioli Tonga from the Pacific Data Hub team at SPC. A question for Nancy. First of all, congratulations on the work you're doing. It's really impressive the comprehensive approach you've taken to strengthen data governance for png ministers, if you can share with us any examples or anecdotes about applying that cultural lens to build trust to help with the adoption of your data policies and framework.
Sorry, can you repeat that again?
You mentioned applying a cultural approach in developing your data governance framework that's really intriguing and really interested in understanding how you may have done, some examples of how that's worked for you, or some of the challenges around applying a cultural lens to developing a data governance framework. Yep.
Okay, that was actually the maturity the data governance maturity level that we use to assess the level of data governance across the public sector. So because we came to understand that data governance is a very new agenda for public sector, and also, there has been a lot of department that kind of don't understand the, I mean, the value of data, so we kind of do duplicate work The same data that is collected, probably by the department, is collected by education. So what we do is, what we came to realize in most of the case studies, is that there is no interoperability among systems for data to join and sharing of the same data. So while we have different government departments collecting same data, the funding is keep wasting on same work, and we have data that the duplicate all the government department so we have worked with consultants, and we came to understand that what they think is we are saying like them, like so we want something that is for PNG, that contextualize how we live, our culture and our regulatory framework, legal system, too. So in with that, we decided, first we said that they argued that it's supposed to be the policy that we were making. Drafting was supposed to be data protection and governance, but we have nothing in place for protection here. So we decided that we need to fix our backyard within the government system. We need to have some guidelines, standards, that's governance in place before we started putting protection. That's how we came to have data governance and data protection and not data protection, because if it were to go to data it would only suit them and us. For a developing country, we believe data governance would come first and not protection, and also it also solve a lot of problem that we're even in, even in regard to decision making, because we have data set has been used for two three years that have been used by government to plan for next year. So we found out that still, it's not enough. Like the public, we don't have proper data in place because, because there is no interoperability of system, there is no one platform, a portal, that all data is there. That's how we became to have data governance, data protection initiative across whole of government approach. Thanks. Also with that, I just want to ask, I'm not really sure if the Pacific Islands, do we have a any national data protection policies in place or act that I mean, it also part of my role, that I would like to know, because eventually we have in cross border data among us. So I just want to know if there's any Solomon Island, Tonga, Fiji, Vanuatu, is there any act or national policies that you already have? I would love to know. Sorry,
from Tonga, yes, we do have a appeal that right now has been drafted and going through awaiting control to parliament on on the data protection and we also working on framework for data governance at the moment. So yeah, it's coming.
Great. Thank you.
Any other questions otherwise? Thank you very much. Can you give another round of applause for our speakers
and all of us
here? Thank you so going for 30 minutes of afternoon tea, and then we come Back here for another session. Thank you Very much. You
recording stopped I
recording in progress.
If I can ask those that are still enjoying the coffee outside. If you can bring your cup of coffee, we need to start on the second last session. You
start the
other session first, but
allow 30 minutes. Okay.
Yes, we will be starting with our Session Four, artificial intelligence.
Artificial Intelligence. We will be discussing about the future of transformation digitalization. Our first presenter will be from the European artificial intelligence regulation, Carolina iwanska, European Center for nonprofit low stitching. Carolina is online, and before we start, we will ask her to introduce us to the to the participant, and we'll give her the time to share a presentation with us. Carolina, hello,
thank you. I hope you can hear me. I'm having trouble sharing my presentation because I'm joining through my tablet, because I couldn't join through the link on my computer. I did send you the slides. Sorry, I am.
We can hear you, but the technical team will be working on putting up the slides. Thank
you, fantastic, and sorry for the for the trouble. It's never you know, an online participation in a conference without some pickups. Thank you so much for the invitation. Thank you so much for introduction. I will try to see if I can still somehow share my slides through my mobile device. But let
me see,
some reason I couldn't join through the computer. It's been tricky. Please let me know if you have received my presentation. Maybe that will clarify things. And thank you all for your patience,
the technical team haven't received the email of your presentation. Would
you be able to
the committee? Is there an email that I should do sorry for the logistics part, I sent it to the committee. Email,
can you check on the
Yeah, yes. They were working on sarees, checking on the email. But since we are okay,
I can start maybe, yes, if we have slides, it's good, it's not I can continue with our slides. That's also a possibility. Thanks everybody for your let me just thanks everybody for your patience with the technical aspects, and thank you very much for the kind introduction and an invitation as well to participate remotely in this event. I wish I was there physically with you. So I'm calling in from the Netherlands, where the European self profit law is based. My name is Karina ivaska. I work as a digital rights advisor at the center CNL. We are a civil society organization that is based in The Hague in Netherlands, and we work on protecting and promoting civil society and civic freedoms, also in the context of digital rights and new technologies. So I primarily focus on our advocacy towards the EU, and this is a technique that I was asked to present on today, to share with you some recent developments when it comes to the regulation of artificial intelligence in the EU, in the context that it's what, it's one of the first comprehensive regulations of AI in the world. And in that sense, it can be, not necessarily an inspiration, which you will see now in a minute where I claim that, but it can be a starting point for discussion of how to regulate AI. Oh, there we go. Thank you. Fantastic. If we could move on to the next slide. Just a little bit of context for the EU artificial intelligence act. So it has been adopted in 2024 in early 2024 it just came into effect in on the first of August. However, there is a period a couple of months before actually start to apply. And it will start to apply in a gradual manner, uh, first provisions will come to effect in February 2025 but the full application is going to is only expected in 2027 so two years, sorry, 2026 two years after entry into effect. So that, you know, by virus AI systems and the powers of AI systems can have time to, you know, to fulfill the conditions this regulation will be directly applicable in all EU member states that So ultimately, EU member states, and the broad aim of this regulation is to really harmonize the internal market on AI within the European Union. So that means, what are the conditions that those companies or public institutions that want to develop, sell or use AI in the EU what are those conditions that they should fulfill while protecting human rights, fundamental rights. So I will be presenting a civil society perspective on this. We were involved in advocacy around the artificial intelligence act from the very beginning, so from 2020, where the white paper was released by European Commission with IDs for how to regulate AI. And we've been very active in the process, together with our partners at the European Rights Initiative, which is a network of organizations in the EU and this is it has been a lot of work from civil society around this dysregulation, and I will share with you what are the key elements of the social intelligence act. Of course, it's only going to be a sketch because we don't have money, but I hope, hopefully it will provide some context into what is the regulatory approach. What are the positives or the negatives of this, of this approach, and what happens moving forward. If we could move to the next slide please, I will the regulation is based on a kind of risk based approach, which means that there is an assumption that certain types of AI systems are more risky to than others, and that risk is in relation to health, safety and rights. So, so that's the context that the regulation is based on, and it essentially defines four levels of risk from starting at the top of the pyramid, so the smallest sort of number of systems that will pose an unacceptable risk. So that means AI systems that whose risk to fundamental rights and health and safety is essentially too high to allow for the development and use of these systems. So in practice, these type of systems are prohibited in the EU, they cannot be developed. They cannot be put on the market. So they can solve in the EU, they cannot be put into use, however, they can be exported. So this is, this is the the tricky part. They still can be research can happen around them, and the EU law will not apply externally. So export controls are not part of this approach, which is problematic, because if you know we assume that the risks to fundamental rights are too high for a specific system. We then, does that assessment differ in other countries? But that's but that's a side note. Then the second category is high risk. So these are the kind of systems that are, you know, deemed to have high risk fundamental rights, and then they essentially will have to fulfill certain conditions based on which they will be able to be put onto the market and sold and used. The two other categories are tests that so called limited risk, and they only have transparency obligations under the AI act. So for instance, to inform people about operations of these systems. One of these systems can be a chatbot, for example. So chatbots have been classified as limited risk, and essentially, providers of these systems, they have to inform the person that is interacting with the chatbot that it is essentially an artificial intelligence system. And the last category is minimal risk, which is the vast array essentially everything that is not identified as unacceptable risk, high risk or limited risk is deemed to be minimal risk, and they're essentially no requirements apply. The AI app essentially only applies to the two top categories, and to some limited extent, to this limited risk category. So now I will go through some of these categories to kind of highlight what are the types of AI systems where types of obligations are there. So if we could move to the next slide, please, and that concerns prohibited AI systems. So what systems are have been considered unacceptable in the AI act. So this is not an exhaustive list. I just wanted to flag to you which with type of AI systems we are dealing with. Essentially the first one I would highlight is real time remote biometric identification, so that could include facial recognition, for instance, and that is would be prohibited in publicly accessible spaces, however, only for the purposes of law enforcement. So in case, a private sector operator uses facial recognition in a shopping mall, for instance, that is not included in the prohibition. However, an important caveat here is that other EU laws do not stop to apply. So for instance, data protection law still applies, and there have been many cases already that facial recognition is prohibited under the GDPR, under data protection law in the EU when it comes to the private sector, which is the sort of justification why only these systems have been included as prohibition, as prohibited. However, there are exceptions to this prohibition. Essentially when the it is necessary for the targeted search of victims of crime, prevention of a terrorist attack, search of the perpetrator of a crime. And these are quite broad, unfortunately, and they are quite far reaching. So essentially, law enforcement, police can say that you know, they need to use those systems to prevent terrorist attacks, and in that case, national legislation can authorize the use of such systems. So the actual question of how effective this prohibition will be in practice is, you know, is very questionable. Some other some other prohibitions include emotional recognition. So systems that prepared to recognize emotional states of people. However, that has only been limited to the area of workplace and education, so that would be only prohibited in workplace and education and for instance, not what has been our criticism throughout the legislative process, not in the context of law enforcement and migration, which are arguably the most risky areas of use of such systems everywhere, there are actually already documented uses of such systems which have brought a lot of harm to people involved. Another type of AI system that's going to be prohibited is predictive policing. So any AI systems that conduct sort of individual risk assessment of whether someone will be a criminal, or whether they will be offend. Or, you know, you can think of the famous
kind of the infamous system from the US compass that was predicting the risk of recidivism, that was highly discriminatory. So these are the type of systems that will be prohibited. There is also biometric categorization. So kind of links to identification, it's about putting people into certain categories based on their sensitive characteristics and biometric data. So for instance, you know, using technology that is not necessarily recognizing people faces, but it's, for instance, recognizing the way that they work, which is also biometric data, and putting them into, you know, risky, non risky typology that is also banned, with exceptions. And then finally, social scoring. You can think of the social scoring system in China as an inspiration for this one. It's essentially a prohibition of, kind of evaluating people's trustworthiness and creating social scoring about, you know, whether these people will be able to access education or benefits or, you know, public services, so that will also be prohibited. So as I mentioned, these are, our assessment is that it's great that those prohibitions are there. That's already a really good step up towards the status quo. However, the fact of those far reaching exceptions make it possible to actually authorize most of these systems in practice of how control, how oversight will look like, whether this is not going to be abused, there are no prohibitions in the politicized area of migration, which is a very political topic at the moment in the EU and that has essentially been removed from the area of prohibitions. So for instance, you know this remote biometric identification, it can be used technically in the migration area, but that law enforcement, so this is highly problematic for protection of rights of people on the move. And then there is also a bigger issue is that there is also a very broad exemption for national security, so none of these prohibitions will apply in the national when national security is invoked by member states, and that is very difficult to control. So it remains to be seen how effective these prohibitions are going to be in practice. They already kick in in February, so they will start applying from February 2025, and a civil society will be monitoring how this is this is being put in in practice, and we can move on to the high risk systems now. So next slide please, and just to highlight what are the type of users that have been classified as high risk systems. I will not go through all of them. You can just see what type of assistance have been included. Essentially, they are linked to many of them are linked to public services and areas where human rights can be most involved. So education, welfare system. So for instance, access to benefits, law enforcement as well in terms of systems, are not prohibited. So for instance, geographic predictive policing, so assessing whether a particular neighborhood is, you know, high risk in terms of crime. That is high risk. It's not prohibited, despite calls to prohibit it by civil society, as I mentioned in migration, those systems are being used to, for instance, assess visa requests or visa applications, asylum applications to conduct assessment of risk of individuals who are crossing borders. That's high risk as well. And then these are, mostly, as you can see, kind of public sector uses in terms of the private sector. And some you know, if private companies use biometric systems lawfully, that would be considered high risk also employment area. So if systems are using useful recruitment or worker management that is also high risk and also in actual sectors. So credits, credit scoring, as well as assessment of insurance premiums. And this list is updatable. Every year the European Commission can update this list based on emerging risks, which is good because it makes it sort of flexible and adaptive to emerging risks. And then what does it mean if a system is categorized as high risk? So if it's falling into one of these categories, then it means that strict regulatory sectors decline, and these can be distinguished between providers or developers of those systems and deployers or users. So providers conduct a conformity self assessment. So it's based on self assessment, they essentially have to check the risks, conduct an impact assessment, implement mitigation measures for those identified risks, check for data, quality, prevent biases, maintain technical documentation, kind of ensure high level of accuracy, robustness and cybersecurity. So you could argue that these are, you know, the sort of things that any responsible AI provider should do anyway, but it's being essentially mandated on them. And the important thing is that also the deployers. So for instance, if there is a company that is developing a system that is assessing whether people should get or not benefits, and then that system so they have all these requirements that I just mentioned, and then the deployers. So the public authority, for instance, the municipality of Amsterdam, let's say, if they're using this system in practice, they also have to conduct a fundamental rights impact assessment before deploying the system. So there are additional checks for deployers to make sure that they don't violate people's rights. There is also a Transparency Register where both public sector deployers and all providers have to register their systems and provide a certain level of information about how these systems work, which can be a tool for civil society, for activists, for journalists, for researchers to actually check you know, what are these? These systems are being used and in terms of enforcement, there is going to be a post market surveillance by an authority. But we are going to touch on that in two slides. But last thing I wanted to say here, if we can move on to the next slide, is to talk about a special category of AI systems that has been included that does not fit into the risk pyramid, as I, as I showed you, which is the general purpose AI system. So think, you know, chat, GPT, other language models. So that is a category of systems that doesn't have an intended purpose. So you have seen in the past, in the previous slides, that all these systems are about, you know, they are intended to be used by law enforcement for a specific purpose. And these are the sort of systems that don't have a predefined purpose. So they can be used for multiple purposes. Chat, GPT, I think, is the example that many people will will know. So essentially, they also will have some obligations under the AI act, from, you know, maintaining technical documentation, implementing a copyright policy, ensuring that data that the system is trained on doesn't violate copyright, some publication about the sources of information that are used for training. And then the biggest, the biggest general purpose AI systems like, chatgpt, they have been classified as so called systems are posing systemic risks, and they will have additional obligations. So they will have to identify and mitigate risks. They will have to document and report serious incidents like violations of fundamental rights, for example, and ensure cybersecurity. So there are some there's an extra level of protection on those biggest, most impactful systems. So if we can move on, because I know we don't have much time, I have two more slides, we can move to the Q and A. I just wanted to briefly touch on enforcement. So what is the idea for the enforce? You know, how do, how are these rules actually going to be enforced? As I mentioned, a lot of it is self assessment. So it's actually up to providers and up to deployers to say, Oh, my system has these sort of risks. This is how I'm going to prevent them. This is how I'm going to mitigate mitigate them. And this is very much embedded in the sort of market surveillance logic, because this law comes from a long tradition of product safety legislation, you know, think chargers, you know, light bulbs. So it's really up to the, you know, the company, to ensure that certain standards are being maintained. So there is no straightforward complaint mechanism for affected people, which is a big criticism that we have a civil society. Because, you know, if you have been subjected to a system that should be prohibited, you essentially have no recourse unless the authority takes the initiative to, you know, to actually enforce the law, which is highly problematic. So the but the So, so there will be attempts, I guess, by civil society to, you know, establish pathways of communication with enforcement authorities to make sure that there is a way to flag violations. So enforcement will be divided between, you know, this is, I will not go into Mexico here, but will be divided between the EU authorities and the Member States and the European Commission that there is a new entity called the AI office, which will be enforcing the rules for general purpose, AI systems, so think chatgpt And on the national level, it will be market spans authorities which will be identified within a couple of months in each country, and They will essentially
enforce the law visa. We both providers are based in discussion and or putting, you know, this AI system on the market in a specific country, say in France. And if the French police, for instance, uses an AI system, then also the French authority will be the Enforcer. An interesting thing that I think is worth it's good for inspiration for like legislation elsewhere, is that there are two bodies that will include external expertise, and that's the scientific panel and the advisory forum. The scientific panel will advise the AI office when it comes to general purpose AI systems. So really, sort of like most, let's say, cutting edge, like new types of systems, and this is going to be composed of independent scientists and experts. And the advisory Forum is a consultative body that will also compose, be composed of, you know, company representatives, but also civil society. And that is an important kind of win for civil society that we will have a sit, seat at the table when it comes to advising the commission and the other authorities in terms of how to apply the AI act. If we can move on to the last slide before, before questions I wanted to you can probably, probably gathered for my presentation. Maybe, sorry, maybe just the previous slide for one minute. If you, if you could gather from my presentation, the AI Act is an important effort to regulate AI, but it's far from a golden, golden standard. It has way too many loopholes in terms of protecting fundamental rights. It does not prohibit, you know, systems that shouldn't prohibit affection recognition. In all cases, it is highly reliant on self assessment rather than, you know, true regulatory sort of oversight. It doesn't really have complaint mechanism. There is a broad national security exemption. So there are a lot of things about the AI act that are very flawed. So it's a warning. I think a civil society, whenever other countries think are thinking about AI regulation, is at least, you know, to avoid these pitfalls. So last slide please. And this is indeed fitting into a part of the larger regulatory trend. So the EU often tends to have a lot of influence on other regions in terms of how new technologies are being regulated. And this has been the case already. For example, you might have heard that also Council of Europe, which is a, you know, it is broader than the EU. It includes more more European countries. Also those are not part of the EU has adopted a framework convention on artificial intelligence, which is very international treaty on AI, and it's open to signatories from around the world. So there is a chance that it will become a kind of global regulatory standard. And this is highly inspired by the AI act. It's based on the same logic. It's weaker than the AI act. So it's kind of worse than the AI act, because there had to be, you know, compromises with participating countries. But, yeah, it's another effort to regulate AI on the global level. And we have also seen, you know, a lot of national level legislation in US, China, Canada, Brazil, Costa Rica, others I'm not aware of. You know, maybe there have also been developments in the region as well. So this is really a trend that we are seeing also on the UN level. Recently, there has been a UN General Assembly resolution or occupational intelligence that is also, you know, it's un so it's a relatively high level, but it's still, you know, showing a political will to to, you know, rein in these technologies, as well as, you know, the global digital compact, which is under negotiation right Now, and it's going to be adopted soon, that also has a chapter about artificial intelligence, and aims to create a standardized approach to making sure that the systems are being developed in a responsible way. So I stop here. Thank you so much for your attention. Apologies for the first hiccup with the technical issues, I wonder if there are any questions that I can respond to. Think I saw something. Thank
you. Carolina. Do we have any questions in regards to the presentation?
I do see a question in the chat, so maybe I could read those out and respond first,
oh, there's questions in the chat.
So I see, I see a question from Brook, two questions. So the first one is, what does remote mean when it comes to remote biometric identification, and whether it concerns, you know, the type of processing, where is processed. So at this thank you for that question. Just to clarify, the remote means it's not related to the type of processing, whether it's on, you know, Edge cloud or, you know, somewhere else, it's related to the actual physical presence of devices and people who are being under surveillance. So if the if it's happening at a distance from the person involved. So, so you know they're not for instance, you could the non remote identification is when you are passing a border and you are being scanned your face at the border. So that is not remote identification. The remote is when you are walking the street, and the camera is like far more away from you and you cannot see it. For example, you don't know that it's there. So that, that is what remote means. In this context. It's really linked to the physical aspect of surveillance, rather than where this data it doesn't matter, essentially, where this data is being started process that is not relevant in this context. And then I think the second question is also linked to to the same to the same topic, but it's related to the GDPR. So the GDPR has long had the right for decision to be made by a human, and whether that also applies to remote biometric identification, that is an excellent question, and you're completely right. So the GDPR, as I mentioned, doesn't stop to apply. It's still so maybe, just for those of you who are not familiar with this provision, the GDPR says that a person cannot be subject to a decision that significantly impacts their life, the legal situation, or otherwise significantly impacts that is made, that is fully automated, that is made fully by an AI system. There has to be human intervention, and this is the case here as well. So nobody can be arrested just based on the remote biometric identification system or no hit. Even in the AI Act, there is a requirement for police officers to verify that information. There is, even if I recall correctly, a double verification. So there is actually a higher threshold. They are, you know, required to actually have two police officers essentially independently assess the the result then any decision about starting an investigation or arrest or any other consequences, it has to be a human decision. So that is still, still the case, and real time means that that assessment is being done, you know, essentially real time, that whether you know the cameras, the facial vision cameras, they are processing that data in real time. There is also the retrospective facial recognition. So for instance, the analysis, and that is considered high risk is that banned, the analysis of CCTV with biometric software, so already captured information, and the analysis of that is, it's not real time, and that is not prohibited, which to me is, you know, is a problem, because it doesn't matter if it's in real time or in post remote biometric identification, the consequences are same. So you know the incredibly discriminatory violating the essence of privacy. The European data protection supervisor, for instance, which is an authority for data protection on the EU level, has also had calls to the EU institutions to ban both biometric identification in all cases without any exceptions, both in real time and retrospective. It hasn't been done, unfortunately, but this is where the GDPR still continues to apply, and we are a civil society going to be using, still these provisions to challenge these systems. I hope this answers your questions.
Thank you, Carolina. I will ask the audience if we have a question here for you. Yeah. We also would like to thank our participant online for joining us and for also participating in this session. So Jonathan Royce of New
Zealand, I'm I'm wondering if there are opportunities you know of for Pacific regulators to learn more about these regulations. They seem very, very complex, and I think the regulators might might benefit from from some deeper than one hour education in it,
for sure, I definitely agree. It's hard for me to say, as someone who's not involved in, like the intergovernmental exchanges, to say, what could be those fora? I think, you know, the IGF is a good forum for that, for showing lessons, essentially. But I also know that there is, you know, sort of interest and increasing trends to organize so called AI summit by different governments. I know the French government is money to organize an AI summit next year. So this could be good opportunities to exchange, although that is more on a political level, I think, but however, on the kind of, I think, you know, the commission and AI office, they do have a sort of international program in terms of, mostly, you know, promoting EU regulation. But it's not even now about promotion. It's really just about exchanging and understanding how it's going to work. So I would say that regulators, should, you know, speak to each other through these kind of existing ways of collaboration. I know, for example, that many data protection authorities are going to be designated as authorities under the AI act. So, and I know that already a lot of international cooperation exists for data protection authorities. There are, you know, global data protection conferences specifically destined for for the authorities, and that's a good forum for exchange. And I think you know if, increasingly, countries around the world adopt a legislation and adopt also authorities and regulators that deal with it. I think New fora will be needed and will start to emerge. This is, you know, we only scratched the surface today. And as it's this law is, you know, I had to read it multiple times to fully even understand what is happening. I'm happy to, you know, forward to the organizers some resources that I would recommend as a starting point, but ultimately, I think some cooperation fora will have to be developed. I would say IGF is a starting point the global IGF is a starting point for exchange. And we'll also have to see how the AI Act actually works in practice, because the very new regulation is very vague in many aspects. You know that many of these obligations are not operationalized, and actually will take a lot of time. Will take two years to operationalize them, to develop standards, to develop methodology, and that is something where I think will be worth following, because this is where the devil will rely in the details in this context. So yeah, I would, I would encourage you to follow also civil society organizations, if you're interested. That we always have a lot of materials, educational materials, mostly towards civil society, but I think it can be also helpful for other stakeholders. So sorry, I don't have any more concrete answer to this, but I think we it is just a matter of time that some form of collaboration for our will emerge.
Catalina for that explanation, but we would like to ask if you can stay online. We would like now to ask the second presenter Thank you, to come on stage, and maybe after that we can receive more questions from the participant. This afternoon, our next presenter and his team will be presenting on the usage of AI in the Pacific is well known in the academic fields. I think it will be best if he introduce himself Dr Sharma. I don't want to lie to the colleagues here and online, but please, the floor is yours and your team.
Thank you, savvy Bula and Pacific. Greetings to you all. Good afternoon, good afternoon. Sorry. Being a teacher is always a teacher. You have to ask twice. You know, we say good morning, good afternoon. Nobody says it. Everybody's busy with their smartphones, etcetera. So the second time around, it makes some sense that people are actually responsive. I'm Bibi Sharma from University of the South Pacific, and we have our colleagues. They are not in my team. They are in their own separate teams, by the way, and they will join me in this presentation. They can come in whenever they want, and they'll they'll speak about, basically the use of AI in higher education and how we are planning out to do things in the future. We played rock paper scissors and I was picked so I'll make a start.
No, oh, okay, right. So while we are waiting, remember, if we are teachers, we don't stop you know whatever is happening. We'll just keep on talking, whether people are listening or not, it's still fine. So yes, I think this AI is bringing to us huge potential, if you just look at it. There's so many things that can happen in the past week, like what, what we had in terms of ICT earlier on, 1015, years on. So I think now we are progressing to AI and like Carolina already said, we still don't know its potential, but we have to be prepared. We have to be prepared like there was a visit one month ago from Dr Ruman Choudhary, the US science envoy, and she said, basically that we are experiencing now AI wave in the in the Pacific. So I'll start off with that. Am I able to start now? No, I think no, the slides have been sent Yes. Would you like to say a few words? Introduce yourselves.
Okay, thank you. Professor Sharma, my name is Ashish dosh. I'm from Fiji, and I'm working at the Fiji National University. We wanted to, actually, we had two presentations, so we are representing the two universities in Fiji. So we decided to do it together, right? And with me, I have my colleague, Miss salah. She's in a manager in applications in the its at Fiji National University. And with myself, I have just been there. He is joining us right from PG at the moment, and he's an instructor in it. So the challenge here, while they're still setting up the slides we have so AI has become a huge issue in terms of education, right? We are still planning on how much AI we can use. I mean, what, what AI is in Fiji, I think the only thing if you say AI, okay, people would know, okay, only chat GPT, right? As educators, we want to also tell students to use AI, but we sometimes want them to refrain and use their brains right, while they can like, not for using for assessments, right, not copying directly, but as I think Professor Sharma and USB and fnu, We are working and trying to set up guidelines. I mean, we as universities, we are also in the dark of how we are going to plan this out, how these things will portray right? And that is what we want to have a discussion on here today. And that is what we want to talk about, professor in
apology, apologies, this is where law enforcement comes in and trying to resolve this issue. So if we can bring mister Jasbir online, and he may be able to start with this presentation, since we are running out of time, and then we'll close it off with Doctor sarma, please.
I mean, with the presentation by Catalina and the work that they have done in AI, we are still learning from what we have missed in the past, and now we have been put into advanced technology in the use of AI. I'm not surprised that some of us are lost and sleeping at the moment. I Yes.
The computer is confused. Yes. Mr. Sharma, please. I think we go back to the order of presentation.
Thank you. Survey, yeah, welcome back. Audience, thank you. I'll skip some of these things because I feel that we are racing against time, but this is basically what we will do. We'll tell you a little bit about what our investors do. What is our target in terms of, AI, some of the gray areas which I feel we should be solving together, and then some of the solutions which we can deliberate upon little bit later. Ai, if you look at the potential, especially in the Pacific, it's huge. And like I was saying, this is basically, if I try to think of something like this. Then in Fiji, we have this season in April where we get lots of nettley or anchovies. And this would be buckets load of not buckets drum load of anchovies coming to the shores. And also it's like anchovy waves after waves coming to the shores. And I feel that AI is basically the same. We have this AI waves coming to our shows with lots of potential and lots of uncertainties. We do know, actually, what is happening. We don't know what is right how much we can for example, Jovi's, we didn't know how much to harvest. We harvest so much that we could not control we let it rot. Okay, so it has its plus and minuses, which we probably can discuss little bit later. And moving on, if you look at higher education, then again, there's huge potential. If you look at how we can use AI for our teaching, for our learning, for our student support, for student retention, and for research. These are five arms, and my colleagues will agree, these are five major arms, which we have in the higher education. I'll not look at the examples for now and again. If you look at what is happening in the Pacific? I think we all know that there are many pockets of excellence. There are a lot of work being done in this area. If you look at AI, even cyber security, there is lots of cases, lots of people who are actually working in this if you see, for example, UNESCO, that was Dr lib Wong, we have Australian government that has committed 39 point 9 million in his last budget for AI work in terms of our competencies and regulatory measures. We have Ministry of Education, Fiji that are trying to look at having AI implemented in the schools in the primary and secondary levels. So these are some good cases, some good work that is have started, actually. And I go back to what doctor Ruman chod was thinking that there is potential of us using AI to address our climate resilience and mitigation. So see why is working in that particular direction. And and people are talking about how journalism, journalism can use AI, because we know there are lots of fake news going around, deep faking if you must have had all these things are happening because of AI. So there is this dark side of AI, which we have to tackle as well. And then we have, again, many small projects that are happening in the Pacific. For example, insurance claims, how it could be fastened, how the payment could be fastened. And like I said, when there is lot of data, we're talking about, AI talking data. And we talk about data, there is cyber security. And again, in the Pacific, we have got many, many cases that have come up, you know, probably more, many more cases than me, we have cyber attacks. In Fiji, we had this vanwatu ransom. We had the Samoa Cook Islands. And just recently in Fiji, we had the M PESA, where people just took money from your M PESA account. So there are lots of things happening, which means that there's a role which we all play, including the higher education, fnu and all the regional universities as well as the national universities. Again, this is just for invest of the South Pacific, but I think this is the same for all the other investors as well. You guys can chip in whenever you want. We had, I just basically tried to say that there are three different clusters, three different levels of where we use AI at universities of South Pacific, there is the institutional level, which I'll discuss a little bit in detail. There is a service level as well as we're going to the cause level. If you look at institutional level, then what is happening across university? Well, there are certain things which we have just started. We have started to have our policy, our ethics policy for AI. We have started to introduce AI ethics as well as use of chatgpt in our generic course. Generic course basically means it is mandatory for all programs, so people, at least, will know the ethics around AI, how to use AI, what is good, what is bad. We also have some documentation for the facilitators on what to expect from students and what how they could use AI for themselves. So these are things that are happening at the higher level. We are also, and I'll talk about that a bit later. We are also trying to introduce some programs in AI. This is something which I think is happening in all higher education institutes where we are trying to improve the performance in terms of HR, you guys have the same would you like to discuss that?
Right? What specifically, in a fnu again, what they trying to do with the the HR, right? They've, I think the first things that fnu is trying to do is, at the moment, we have got a new PMP system, okay, that just that has just got into place, okay? And we have been able to, I mean, kind of, actually, at the first we are trying to get our things right, at the moment, right by introducing a good appraisal system, a PMP system for, for this, for, if you knew, I would say, I mean, it would be a very first step in, AI for HR, right? They're still exploring their ways through it, because USB, I think, is more into with inspects of HR, right? But we are still trying to explore that area of getting our things right. Now, at the moment, there's still, I mean, Salah would be able to explain more end
they've actually, we've just completed our implementation. So currently, right now, I think we're trying to customize and telemedic to suit our university. But just to add on to professor, because fnu is also working towards implementing and designing an AI policy. So just early last month, this is actually led by our learning and teaching unit working closely with our School of Mathematics and computing science. So just early last month, they conduct conducted a student survey, and very shortly, they'll also conduct one for for staffs in terms of supporting our supporting departments. So with the data collected, this is to assist them as well in terms of designing a policy. Maybe Savi can correct me on this one, but currently we don't have any AI policy at the national level. But this is something that, you know will be interesting and also crucial, that that we are in need, and maybe from there as a new university higher education and in, you know, organizations we can always benchmark to collectively work with, because currently we don't have anything at the moment, yeah, thanks, thanks
a lot. See, this is teamwork. Who says University, and if you are fighting, we are not, we're working together. So coming back to our HR, yes, I think there are many things we can do. And the presentation before that was quite, quite good as well, because they basically talked about there are risk for example, when we are trying to scan, we're trying to filter applications and obvious, good person who has got fabricated the grade and fabricated certificate. So how can AI find out manually, may be possible? Ai, is it possible? Well, there are risk involved. It may be possible. So there are, we are still learning. AI is also learning. And also, I think there is a very high turnover of employees you might have seen in the Pacific. Why? Because everybody is going to New Zealand. I was telling you the EN, I said, Ian, everybody is going to New Zealand. Ian said, Well, we are complaining as well. I said, why? He said, Well, from New Zealand, everybody's going to
Australia. So yeah, this is a cycle. But yeah, there's
a problem, and I think we have to know which are the areas being targeted, so that we are proactive and we are able to carry out the staff development, professional development, and also to keep the staff morale high. There's something which we we we urgently need to do in terms of research. We have carried out a number of research we are working in many, many areas, either using AI as a tool, or researching on AI. I think it's important to research on AI as well to find out our perspectives, to find our attitudes of using to find out our readiness sometimes and like, like, Ian, okay, I go back to Ian every time. They also said about one laptop per child, it failed. Why people were not ready. People were not trained. The same thing with AI. If we just bring in AI and people do not know they are not ready, same thing will happen. I'll tell you another story. And this was my first incident with AI. I typed in, I went to chat, DPT, I said, Okay, where's BIPA Sharma? And it said, BIPA Sharma died in 2016 so I don't get along with AI, by the way. I don't like it. So yeah, there are many options and many things we are doing. And I think this is another place where we should be collaborating. All of us should be collaborating. When we look at how we can use AI in climate resilience, in mitigating factors. How can we do that? What are we able to do? And I go this time to Dan, when Dan was saying, and I think Keith was saying as well, Dawn, sorry, Don. And Keith were saying that this prospect of telemedicine, telemedicine is the old concept, 1015, 20, old concept where we used to have these tele appointments, but now telemedicine is different. It has transformed to include diagnosis. It has transformed to include the face level of treatment. And that is what we need for people in the remote areas. We don't have doctors in the remote areas, in all the remote islands, we don't so how do we get the treatment? How do we look into this healthcare may be service like like an app. How do we do that? But again, somebody outside was telling me they will not accept. All people will not accept. There's still one face to face. Well, okay, so this is where we bring in the awareness. We do some research so that whatever interventions we have make sense to our people. I
so this is basically what I was, I was going into trying to see how we are able to use our AI in ocean and climate management. We are talking about our our stocks, fish stocks, ocean resources dwindling. Why? Because of overfishing, we have to use AI to know, okay, right? This is how much stock we should have ensure so that we don't run out, because Pacific cannot survive by just conservation. No, we need to balance it with economy, right? We need money as well. So we have to strike that balance. That balance can be found out those three, those thresholds, can be found out using AI. I'm pretty sure we talked about the cycles of some of the fish in Fiji. What was someone caught? You cannot do that. You cannot harvest in November up to April, maybe November to April. You cannot, but due to climate change, the breeding cycles have changed. Have we researched on that? No, so there is no use having the pain during that time. Sorry, I get upset little bit.
Yeah, I think just to add the main issue here is, is we need to find that line, that fine line is how much AI we can use, right? And where is it, where it can be an advantage and where it could be a disadvantage to us. I mean, like things like this, it's, it'll give us more in depth, right? It assists, like we as educators. I mean, I confess I use AI, okay, right? I mean, we all do, and I do, I encourage my children to use it, but what are the ethics of how much we can use, right?
Thank you. And just to add little bit more on that, I think this is again a very important subject for us. When we talk about the E waste system, we again survive on tourism, but tourism has to be sustainable. To have sustainable tourism, we have to ensure that we have E waste system. E waste is not electronic waste, but basically to use ICT to control our waste system. Maybe we can discuss this further later on, when we have time.
Dr Sharma, if I can add on to that. Yes, the things that we are discussing right now, they come at a very critical time, and so did our ROX indicator survey that we that a lot of stakeholders present in that forum would have been a part of, and through that forum, we were able to voice our concerns in terms of privacy and ethics with the use of AI. And since the Pacific Region is on the brink of introducing laws, major laws, major law reforms in terms of AI, use of AI privacy. So a lot of our inputs have also gone into the Romax survey, and through which a lot of decisions will be made on the type of laws and areas those laws will govern will be introduced. So yeah, so we have done that as well. We're sort of looking at putting it out there as well. Thank you.
Thank you. And moving on to the service level again, this is something which I believe most of the investors are doing. They're trying to update their Moodle, trying to look into the eternity in to ensure that they have this early warning systems. These any learning systems, come out very easily. If you look at the land analytics and learning analytics, which, again, the pattern can be, can be brought into light through, through AI, our students, actually. If I look at this one, our students, do they use chatgpt? Yes, all of them, isn't it? So I received this email from students. They say, most honorable, respectful, rightful, sir, as soon as you get that line, you know, this is chat GPT, you know. And the email is about two to three pages long, and at the end, there is one sentence saying, please extend the deadline of my assignment so our students know how to use jgpt. It's not a problem and but I think what AI brings into this game is trying to use this Nana analytics to ensure that we provide more personalized learning for them, more adaptive learning for them. You know, this is what we we want to do, because not everybody has the same learning style pace. So how do we bring AR? Well, we could, we could again, Anju, you are exercising, and again, we could make a bringing material specific for a learner. You'd like to add something? Chris, yeah,
I think for materials right, specific to a course. I mean, I, if I, if I have to take online class, right, and I have to drag it to three, three hours, maybe, I mean, I generally just go to chat GPT and ask, okay, what, what are some of the, I mean, activities or workshops I can do on that particular day, okay, specific to the topic that I'm covering. So me sitting, and most probably, I mean sitting there and looking for activity, thinking, so it just gives me in two or three minutes, right? I mean, what I need to do otherwise, just, I wanted to add on the the projection that Professor Sharma has said, Okay, each universities in university in Fiji, okay, they do a projection of like, okay, year 13, this many students will finish, yeah, okay, after the exams, and we need to get this much students here, okay, like, fnu has this quota, okay, this is a threshold. We need to go. USB has that. So maybe I'm using AI, okay, and we have, like, power, power BI, with the co pilot built in. It will help both universities, right? I mean, getting, I mean, there is a bit of competition there, but you still want to share, okay, but I mean, like looking at a business perspective, it does help, right? I mean, in analyzing and telling you, okay, these are the most, probably many students you might be getting, and if you would, I mean, the university might need to increase the capacity, okay, it can help, also help the university plan things for, I mean, to create a web, bad environment for the students.
Well, I agree with that, because not all universities have the same programs, very successful, right? So I think using these sort of patterns and doing this sort of prescription from Ai, we should be able to division of labor. You guys take some programs. We'll take the major programs.
And this is for the lecturers, again, how we use AI in our teaching. Actually, staff are now being able to create content for their lectures, which frees them from all these other work and they can do more of a flip classroom kind of teaching, yeah, you know. So they have more time to concentrate on their teaching, instead of working on these notes, which is already there. But again, there are lots of other things that come in, like how to do the assessments, how to get the data, and then how to again, get students in certain streams so that we are able to help them in a more personalized way, as well as get a feedback from the students through AI patterns in saying, Okay, this is how we should improve our teaching, improve our lectures, improve our delivery. This I will discuss tomorrow and coming towards the end, this is the challenges which we face, the issues in the Pacific, whether we talk about academic integrity, we talk about transparency and accountability like something what previous presented it. Talk about trying it lacks transparency. Whatever gives you the decision or the prescription. You don't know the process. So you don't know whether to say yes to it or not, how much to trust, okay, and you don't know how there's a lack of, sorry, there's a lack of understanding of the AI conclusions. So you had something to say.
Yeah, sorry. If I can add on to this, onto the challenges from the specific context. One of the major challenges that we are seeing globally with AI is job displacement. But what we might see with a lot of developed nations is that when people are displaced from their jobs, they have a lot of industries. They can sort of train people to in one of those industries which AI cannot really take over or placing is not affected. About Pacific, that might not be the case. We don't have a lot of industries do our manufacturing, and all of those which are labor intensive, where people can go and actually work, but they're not at that scale, right, where people, hundreds of people, displaced at once, can be absorbed at once by those industries. And when we look at our major industries, especially for Fiji, and I'll come back to how this is a university perspective thing. When we look at Fiji, there are, there are a lot of that. We don't have many industries which are at scale, which which employ hundreds of 1000s people. There's, there's your totem industry, which is your manufacturing in those sectors. But apart from that, there's much farming we've seen is declining, especially the major crop, sugarcane, you know, the farming of it, the processing of it, you know, it's, it's a major in the decline. So what we need to consider is how, with the adoption of what we need to see first is that option of AI does not overrun the development of new industries, really, that is what I think, so that we are able, if people do get displaced, we are able to provide employment for them. Because, as it is, we already have a brain drain issue. As you mentioned. We'll see a mini Fiji in New Zealand, and then soon in Australia, and then some other countries, right? So we already have a brain issue. But so with that being said, we want to retain our people with the right skills so that we can develop other people as well, develop our children, youth as well, in that line, right? So what we need to ensure is that takeover of AI, or the introduction of AI, is at a control place, sort of from the perspective of usable tools, usable resources, the list of things that you showed on one of the slides, right? So, yeah. So that is one of the things. And then obviously the universities are coming over here by introducing programs, by introducing programs that allow us to explore new areas, uncharted areas, alright, one of the things I've been seeing as a trend, possibly, is how AI has matured over the years, and what we are seeing now the llms, the large language models, things like Gen AI, they put, I mean, we had, as you said, is not New, right? But the interaction of llms, things like chat, GPT, put it out there to the world that these things exist and how they can be used. And those are huge systems, right? But now the focus of these systems is going to be to shrink those systems with the same capabilities, able to make it specific to a person or a purpose, right? Maybe make it specific to a person of how they can use it for a specific task. Imagine the possibilities of putting such systems on a remote island and using technologies such as maybe AR and VR to teach someone how to build construct. So if we put those skills in the right hands, using technologies, as mentioned, we take education to a new level in the Pacific, because a lot of our locations are remote, isolated, especially the maritime areas and the highlands, right? So, so those are some of my takes in in terms of challenges, before I move on to from this slide. And the other thing, when we talk about job displacement, the slide that you showed on HR department, or the slide that we have in our presentation about the HR department, the guy sitting there tell me how much work will it take to fully automate that HR department? A lot of things that you put in there in terms of processes, the things that it does in terms of analysis of resumes, are bringing out possible candidates for a position, looking at mqas, KPIs. A lot of things are already automated. The only thing the person in the department, or the people in the department might be doing is saying, putting the human touch to it, assessing the human side of things, being able to assess human behavior. So we are already in that process, right, of automating a lot of things using AI. So we need to bring in that balance AI take over and creating new jobs for when AI does take over certain role. So that's one of the challenges that we need to look at from a university perspective as well moving forward. So yeah, thank you.
Thank you. We totally agree you have something to add.
Yeah. I mean, the other challenge we have is it's a very common one in the academic is plagiarism, right? I mean, they stop students not to use chat GPT because it's plagiarism. But what if you tell a chat GPT, okay, write me this paper, give me a referral in text referencing, and then the referencing and everything. So what is that still present pleasure isn't still referencing. It's giving you all the references. Anyone? Yeah, I mean, like we do have inbuilt tenant in that analyze, okay, on the educators, and we are able to see how much AI content has been generated. But for fnu, I would say it does not add up to the total plagiarism count. Okay, you can only see, but it's not added. So if it's 100% okay, AI generated, the plagiarism is just one or two,
actually, 1018, has the ability now. So if you've used 20% of a year, it will, it will tell you 20% of a year separately, it's there with in our Moodle, yes, it's there.
Professor, yes, please.
I think we're highlighting on these two types. One is the where the student sees the percentage, and the other is where the the educator looks at the percentage. So those are not in sync, so it's more the the educator is the one with the report.
Okay, thank you. Um, so we have any short of time, so I'll quickly rush in the last bits, but this is quite important, actually. And I'll we'll have the presentation slides available. Please have a look if you want to discuss something on this is quite important, especially the job, because there is a big discussion about jobs that AI will take over the jobs. No, I will make sure it will create better jobs for you. And again, these were some solutions, but I think this have been discussed already, like trying to look into the policy matters, not just looking at higher education or at a certain level, but trying to look at the regional policies and the national policies in AI, trying to have the governing body, the management bodies in place to have this. And I think it's also very important to establish AI research centers which looks into our climate resilience and mitigation factors. And I will not, I will not discuss this. So this is something which I wanted to show you for promotion, because now we're having a bachelor of artificial intelligence in USP, yes, okay. We
also have short courses in AI in fnu. So if you don't want to study for three years, we just have only one week. Can we switch off? This
was supposed to be for USB only, and I wanted to end with this note, AI is also evolving. We talked about a traditional we now have generative but agentic is coming soon. So for example, if we are trying to, let's say, try to find out 10 restaurants in in Wellington, it will find out 10 best restaurant according to your preferences. But what agentic will do, it will also do the booking for you and do the orders for you, book the table for you. Okay, so it's more like your virtual friend avatar together with you. So things are evolving, we have to be prepared. Thank
you. Thank you.
Thank you very much, Doctor Sharma and your team. I hope agentic AI can also pay my bills. But do we have any questions for the team? Yes, I will ask if we can just limit our questions, since we are running out of time to three. Right? And then we'll end
so I'm over schweder, or as chat GPT also knows me, professor, huge swider from okra University. So what I was going to ask you, so you mentioned the use of or misuse of AI by by students, and one of the roles that I apply in my school is actually that of academic integrity officer. And one of the things that we are finding is that around half of our students now use chatgpt, and they're using it to generate their entire assignments or their entire assessment answers, if it's online assessments, and they have access to chatgpt one way or the other, or they have access to a helper with chatgpt, the answer will come from chatgpt, and there's zero, from what we can tell, zero learning happening on the part of the student. And the problem that we're facing with this is actually quite a logistical one, and the logistical one is that when you sort of go into the second half of a semester and your students are starting to slack off, traditionally, they would not have submitted an assignment. But now everybody submits an assignment because it only takes two minutes to do one, and you don't have to have any idea what's in it. What are you going to do about this? And how much budget is the university willing to put towards marking gobbledygook that's been, you know, created at, you know, a huge volume, and as you say, several pages whenever a student writes to you, you know, how much budget are you willing to throw at people reading and assessing this? And do you see a possible use of AI for assessing whether it's been ready
or not worth an answer?
I actually wanted to say that yes. So this is a very difficult question. I see three different responses coming out from my brain at this time. One is AI tackles AI. So AI, let ai do all the marking which AI has created. That's one. The other bit is to, well, this is what we can do, is to teach our students the ethical views about use of chatgpt. I don't think we can eliminate the use of chatgpt and but if students are able to understand that there is a certain level of their responsibility what needs to be done, then I think that is good. They need to use chatgpt. I think that's good to turn out some help for the students. I think that's very good. But then they will need to know, okay, this is where it stops, and this is what I need to do. The other train of thought that is happening these days is trying to have aI resistant assignments. And people have been talking about different designing, different assessments that are basically AI resistant. I thought of looked into that, and I felt okay, this may be a good answer, but then I feel no, we should not stop from happening. So we should not have aI resistance. It should be more of a hybrid kind of assessments.
Yeah, just, just to add on to that, one of this university I know, and one of these lectures at I University, I will not name, so what they have started doing is they've stopped giving assessments, and they've gone back to exams. So instead of assessment, they give, like, test, short tests, yeah. So, so, so that people, students, they do not copy. So as professor, I said, we can't be running away from Ai, okay. AI is here to stay, and it will. So we have to find ways in order to battle this. The other thing is, we as universities, I think not of us, only in Fiji in the Pacific, we can kind of also in the dark of what are we going to do about is how we're going to tackle about doing this. So I think what my what I would like to propose, is, I mean, we are working alone. Why don't we all the universities in the Pacific, we get together and try to walk a solution out of it.
Thank you. As
a last questionnaire from Samoa, so I'm gonna wear my bike watches hat. So bite watches is AI startup in Samoa. We've been developing a chat pod. We've also presented our chat bot to the National University of Samoa to view what their concerns are. So the AI is a threat, but it is an alarming threat that maybe it can wake up academia. Is it time to change the model in which we teach, because in most academia, we are testing how much you know, but not testing how can you use the knowledge to build solution? Where from? From our perspective, as a AI startup, we want to use AI to be augmented like it takes part of what the humans cannot do, and use AI to do what humans. Takes part of what humans cannot do, and use AI to automate the knowledge gap. For example, we talk about using AI to build courses, so we can build courses, but the ethical side is still with the human so we can have a hybrid model. So whether it so, the question that I want to ask is it time to reform our education system across the Pacific to help AI innovate the minds for solutions. So one of the tests that we presented on at the National University of Samoa was we took real information data from the samo district Bureau of Statistics in terms of market trends, and then we gave it a problem. We had the information of market trends, and then we say, I'm a banana farmer. How do I build my company to meet the priorities or the mismatches of the market trend, and it produced the result, which means that it augmented what took months into a few minutes. So I think this is where we can ask the question, do we need a reform in our education system across the Pacific
Talofa brother? And thank you for the question, I think in in some ways, you have basically answered the question yourself by saying that we need a reform. We need a reform or total reform. You know, we have to look at the curriculum. Change the curriculum way of like we discussed the previous question. Change the assessments. Change how we assess our students. How can we retain the critical thinking factor, the creativity factor, which is now missing, because if total dependence on AI means that these are gone, but then AI is there to stay. If we look at the traditional ways of AI, if we have to go to a university site, website and try to find some cause information, it takes bloody one whole day, whereas chatgpt can show you, because we have developed the chatgpt to find out the cause information and the second level to, for example, the other thing, which was probably answering the previous guy's question answer as well, which was talking about the intelligent tutoring system. We have our people who are working for six, seven hours, and that's it. But if we have this intelligent tutoring system, they can provide students answers, not answers. Okay, help 24/7, in a course, no. But then again, we have to develop those things. It doesn't chat. GPT can make up things we should be wrong. So we have to feed it. We have to keep on making it learn. Example, if I type in my name now it doesn't say I'm dead, it says I'm I'm somewhere. So it's learning as well. So chat, GPT, is there to stay? Ai, is there to stay? It will learn. But we have to reform. We have to collectively reform based on the evidence. Yeah. Thank you.
Thank you. Thank you very much, ladies and gentlemen. And I thank the panelists for this session. Catalina, if she's around, I think she has left us. But to the panelists from the two universities in Fiji, thank you very much for the informative session. It has given us some assurance that our academics are having a good grasp of AI technology, and hopefully that can bring them closer together As a family. Let's show our precision to our No
thank you, Professor, Shaman team, I believe it was an interesting discussion that could go on. And we have come to our final session today. It's on internet resources services in the Pacific Islands. We have three presenters, and I'd like to call them up. We have Russell Decker from upng. We have Bruce Duncan from Alda and jampika from ICANN. So those would be our last presenters. I believe we can. They can give us those presentations within the hour, so you're not going to stay too late, but I want to encourage you to stay with us until we finish this session. So thank you Russell.
Thank you MC Andrews and very interesting previous session. Thank you, doctor. And before starting my presentation, please stand up. Please stand up. Just make a relax. We are sitting too wrong.
And say hello to your neighbors in your language or abino, turlo, Bura, whatever.
Please take a seat.
Okay, let's start my session. I think I'll finish just 10 minutes, because I know this AI is a 21st century, or even 22nd century, and back to my era, 20th century. So my name is Rasul dekarada. I'm originally from Japan, but being Papua New Guinea as a cctfd manager past three years, but I'm living in Papua New Guinea 27 years. Yeah, already captured my presentation or not. It's alright. It's alright. Yeah, no, it's nice. It's alright. I'm just joking. And current, my role is a director of ICTs PNG, University of Technology for your collection, not upng, upng, the universal PNG, Papua, New Guinea. And just this year, I had very sad news. Some, some our directors, my colleagues, by by by a cash, cash for grade. Some academic integrity matters caused by human, not AI sometimes, you know, traditionally, some, some students pay cash to lecturer and lecturer changes their academic marks. It's very common, and not only PNG, but everywhere. Like, for my example, when I was encouraged, I always failed the physics and I gave up my physics subject so I I'm not paid cash, but I offer to my lecturer, I can look after your dog, 24/7, and you can give me some grade, not an A grade, B grade, but in case, past grade. And he agreed he did so I can. I'm guilty to buy, to buy my grade. Not buy, but bruntly, assist. Come on team. Ready or Not yet, if you can't find it's all right, it's all right. I just say, say Beverly, and I can't waste your time. So again, three years back, as I said, 20 years I was teaching one of the education university, and my profession is ICT education, so I won't teach how to use a Microsoft Word, Excel, PowerPoint, it's boring. You can find the answer on the internet. So I always talk to my students. I can't teach what you are, you can find the answer on the internet. So I'm very simple to make some my academic style. What you cannot find on the internet. I can teach you, for example, previous university, yes, yeah, please. Second, second page I can control by myself. Panic. Okay, so again,
not familiar with this button. Yeah, sorry. Again. Back to my truck after I joined to this PNG University of Technology as a director, and in my job description and also duty statement, they are not stating about this CCTV manager role.pg and for the history of.pg In 1991 assisted by Melbourne University, maybe I forget, yeah, Robert Earth assisted us to study that by this.pg domain is early days, but nearly 20 years, no one look after this.pg domain roles. So I think one my friend mentioned every profit process doing by manually. You have to download our PDF form, or otherwise before email pill, they ask to us by a telephone, and we can send a fax to this application form, fill it, stamp it, send it back. And the process by manually, and we are doing Excel spreadsheet, everything Excel spreadsheet, until I joined to this university. So you imagine, as I said, I'm back from 20th century or even 19th century, everything manual process. But after I joined to this university, I found this, this adult PCC TV is a gold mining but unfortunately, 20 years no one touched it. No one did it. It looks like same as other Pacific nations, I believe some, some pashuk Friends, already kept captured this automated registry. But still we are under the under the automated process. So three years back, I joined to this university as a director and the ccTLD manager. I attend the several meeting APTLD, APNIC, also, ICANN, too. And finally, I met jolan. Thank you that I didn't meet you. I think still we are doing money. Thank you, Bruce and rosemary, other partners always meet me again and encourage me, because when I was in this university, I always feel alone. I'm very alone and I'm very sad and but if I travel to here, yes, I'm feeling very cold, because this is a very cold climate, but I'm not feeling alone, because many friends with me, many one talks and the poroma with me. So I'm not feeling alone. So assisting by this strong support by Auda. And first of all, we need to review this policy, because nearly 20 years since 1998 we didn't review this policy include the pricing. And after I found this policy review always my boss, Vice Chancellor, kick, kicked my ass, and he said, You need to increase the pricing first. But I said, BC, just wait. We have to. We have to improve our service first, and then we can review the pricing in PNG. Past 25 years, every price increase rapidly, let's say three times, five times. So you are right. We can increase the price, but before increase the price, we need to improve the service, which means that this automated this registry. So I called for the tender just a few months ago, and some of the registry, registry service providers already applied the tender, but after that, I'm stacking this working visa program. So past a month, I went back to Japan remotely. I asked assist to our guys, but still pending process. So very soon we'll automated this, Doc, PG, registry and this operational side, maybe, but important things is many partnership automated registry is a fine and also we want to introduce this three hour model, and also png DNS ecosystem, and also multi stakeholder model, which I learned from ICANN. And some ideas, yes, I have several ideas. Include, I didn't, I didn't mention this a PNG IGF, which I already talked to Andres this afternoon. And I want to build up this PNG IGF, and also DNS Research Institute, and also important part of our university is a youth development same as this, ai, ai, topic mentioned, so I can do it, I can open the door. But who will enter this room? It's my young staff. It's my young youth. So just few weeks ago, this is a YES Program, but just last month, July, we conducted this hackathon program with our student and the first I introduced this hackathon program to our university. Many of students asked me, What is hackathon? Hackathon means hacking bank account, hacking this student database, hacking personal information and hacking some girlfriend background. No. No, no, no, no, no, no. This hackathon means that you have to identify current problem. Yes, in PNG, there are a lot of lot of problem, but who will fix it? You as us. So we have to identify the problem, and we have to fix this problem with some technical solution, not political solution. We have a lot of political hiccups and the problem, but not this one. So these are youth entrepreneurship program is also very important. Yes, we can produce 1000 graduates a year, but where is the job opportunity, which our professor already mentioned? It's very important. And also, you know, you imagine in PNG, less than one or 2% shouldn't reach to this bachelor degree. Let's drop out. Drop out again. Drop out and how they can survive. Many people must go back to their village, their home, and they can study their own business, like agriculture, farming and the fishing, forestry, tourism also important, but not many job opportunities. Now, many people remain to stay in city and towns. Where is the job? No job. No one talk. One talk means a family friend look after these young guys. No now many street, street children in Port, mostly our capital. Very sad. Where is our job opportunity? Digital is the one with the key component. But we don't need a bachelor degree. We don't need a master degree. But some ideas how we can study about some cartoons, animations. Yes, my elders are want to study in Japan, but because of this, like a content animation or computer graphics. Yes, many things in there. But how about in PNG? How about Pacific? So this is our main aim, my aim, and also currently we are making a good partnership with the University of New South Wales. Fortunately, I made a travel from Tokyo to Sydney, and the one stop over the Sydney nearly 12 hours transit. So I visited. There is a UNSW sitting campus, and I'm very, very innocent. So I asked you the taxi driver, and it's a taxi taxi driver. Maybe crazy. He charged me around 100 bucks from from airport to the campus, but it's all right, it's fine. I visit the campus and to see the setup, and also this team already visited my campus, and we definitely need to academic curriculum review. Currently, we have some academic program under the mathematics and the computing. But these are mathematics and the computing, especially for the girls. Many girls fail over advanced math subject or even physics, so they are not reaching to this math and computing computer science program. But we definitely need to review this program,
and also, I'm engaged a lot of our graduates to our cctfd is a DNS currently I have one admin officer and one technical manager. That's all, but we need to engage, for example, under the other maybe more than a few 100 employees. But how about us? So we have to create more opportunities, and the people making happy, and the people must attend such a regional conference and the International Conference and the PNG, as you know, not small country. It's a bigger than New Zealand. We have a more than more than three times population than New Zealand. Of course, Australia is a big country. As an Asian country is a big but we have to rise up our flag always. And we are not free riders. We have to contribute what we can do, what we can make, a good partnership. We are not poor country. We have a lot of resources, include the human resource, because in PNG, more than 50% population is below 20 years old. Example, in Japan, more than 50% population is above 65 years old. Many old people, and they exit. But our country, PNG, is a very young I'm very lucky. I'm standing in front of you, and I'm the right mix of Asia and the Pacific. If I'm a Pacific and Asia, I have only one or two kids, but I'm lucky. I'm I'm married with a PNG local day ready, and I have a three boys, and the two adopted. It's a typical Pacific culture. So I'm very lucky standing, in front of you and just short presentation, and my apologies if I'm not making a right word, thank you.
Thank you, Russell. I'm good. So our next speaker would be Bruce from Auda. Thank you, Bruce. Bruce is going to present an interesting concept for the Pacific, so I look forward to what he's presenting on behalf of Auda. So a concept about registry. So those of you who are from the ccTLD, it is interesting. It will be an interesting presentation for you in the Pacific. Thank you, Bruce,
thank you, Andrew, do I need to use this? What does this work
as well? See if I get some slides. There we go. That's very good. Okay, so one of the things is to think about the context that a user has when they choose how they get online today. And the reality is people have a lot of choice if you if you have an organization you're selling some sort of product or service, you have the choice. You can use online marketplaces, of which there's numerous. You can use social media. So there's the Facebooks, Twitters, tiktoks, etc. There's directories that are often specific, like if you're a trades person, there's directories for accessing trade services you have in the health sciences area, there's directories that give you lists of doctors and things like that. Then you have, in terms of your presence, it's very easy to get a free email, very easy to sign up for something like Gmail or Outlook and immediately start sort of conversing with with your community. Then domain names come in, and you can have your own website, and you can have your own email address that's branded to you, which we would say is looks a bit more professional than saying your gmail.com and there's sort of more of a sense that it's a real person behind that email address. So there's lots of choices. And then if you think about what we're trying to achieve with a domain name, and particularly a country code based domain name, and I think this really applies to whether you're in Australia or New Zealand or any of the Pacific Islands, fundamentally, the reason why consumers will access a website that's got.au in Australia, for example, or NZ in New Zealand, is in many cases, people want to actually support the local community. And so if you're in Australia and you're buying something or using a service from.au you're trying to support the local community, there's a lot more awareness now of sort of environmental issues. And again, people think about, what's the air miles if I buy something because that had to travel on airplanes to get here, could have just been delivered by somebody on a push bike or a scooter or something, and therefore not, not use much damage to the environment to deliver it to you. So there's a notion, if you buy something local and you're in the local region, you kind of expect it to be delivered locally and not delivered over long distances. You also expect it to be delivered quicker. There's also the sense that there's local support. So if I buy something, a product or service from my local region, and there's some problem with it, presumably, in worst case, I can actually take it back to where it was manufactured or made and get it repaired. So it's that notion of local support, and there's also the notion of local law enforcement support. So if something was to go wrong, maybe you purchase something online and it didn't get delivered, or it got delivered and it was in pieces, and if you try to contact the provider and they didn't help, you've still got a backup of local law enforcement, which you wouldn't have if you're purchasing something from from another country. So these are all sort of various aspects that are sort of built into the notion of a domain name that's that's localized for your country, and you're trying to provide services to the local community, or even make it clear that if I was buying something from, if I was in Australia and I wanted to buy a rugby top or something from New Zealand, there'd be some sense that the rugby top was actually Made in New Zealand. It's authentic in some way. So those are sort of some generic reasons for why people would want to see a country code and purchase a domain name there. But one of the challenges here is it's got to be easy, because if you look at all those other options I had on the first slide, if I want to establish a site to start selling some some product, maybe I'm in the business of making rugby T shirts or something, or printing rugby T shirts that I can just take a photo and load it up on Instagram. I really don't have to do very much at all. It doesn't cost me anything. Likewise, very easy, just to post a picture of, you know, I'm a conference presenter. Here's a photo, just a selfie, upload it to Facebook, and I'm done. I've now got my conferencing business up and going. So that's what we're competing with as country code managers and in this region. And so you've got to make it easy and so and you also got to keep the costs down. So a lot of these other services are free, and so in some ways, we're competing against free. We might be competing against perhaps the cost of registering a business with the government. So there's some other costs. One of the things I got from Russell was, I think his measure of the cost of ccTLDs is how many cans of Coke, and he sort of uses that as a measure, and he's traveling around as how many cans of Coke would it take to buy a domain name in your country? I would probably translate it slightly differently to make sure that the domain name is less than the price of a glass of beer at the local pub. I'm finding that's not a very hard target now, because the price of the local beer in Australia is going up pretty rapidly these days. So we can usually keep a domain name less than the price of a beer, but you've got to keep the cost down. And then another thing about making it easy is one of the barriers that we find when we ask people, why haven't you established a website? A lot of people will say we'd really want to establish a website. We think that's a really good idea, but it looks really hard. They their perception is the difference between creating a Facebook page or a Instagram page and a website is some huge difference, and that they need some expertise that they wouldn't otherwise have. They're not an IT person, and so they're scared of getting a domain. I'm getting a website because I think it's hard. And the reality is, there's a lot of tools now where, in fact, there's artificial intelligence, you can actually automatically generate a website now. So I could say I want to create a website, and I'm based in Wellington, and I sell hibiscus flowers, and I make it available for people that attend conferences in Wellington, and just with a statement like that, a complete website would be built with pictures of hibiscus flowers completely integrated into a payment engine and already optimized for search engines. And that can be done in five minutes. So it's actually not as hard as it might appear. But the challenge with a lot of these tools, there's e commerce tools like Squarespace and Shopify. There's auto website generators like Wix. Often when you go to these websites, they'll actually ask you if you want to have a domain name, and not offer a selection of domain names, but they're generally major names like.net or major countries like.au or.uk partly because these big IT companies have integrated to a few large registrars, and those registrars carry The country codes that they find easy to offer. And so to really allow your community to take advantage of these tools, not only do you want to be make sure your domain name is available within those tools, so I don't have to go and buy it separately, because people would really struggle to buy a service from a company like Squarespace and then manually be able to configure a domain name that they bought from somewhere else they connected into Squarespace. Very few people managed to do that. So they'll go with what's offered as a default, which is probably.com and not because they necessarily want one of these other names. I'd like their local name, but it's too hard to connect. So comes back to sort of making it easy. Likewise, when you're looking at things like email, if you want to establish a professional presence and set your domain up, you don't want to build your own email service. You want to use one of the major ones out there, whether it's Gmail or office 365 but it has to be easy to configure. Otherwise you would just stay with gmail.com because that is easy. I can create a new Gmail account in under a minute, but if I said to you in the room, connect an email address to your own domain name, most people here probably would struggle, especially if you're trying to configure the DNS settings in the sort of the Google application. So you've got to make it easy for the to sort of come in through to service provider like Microsoft or Google or SquareSpace. Select your country code as an option if you're from the local community, and it should be no different to getting a domain name like.com or.au i so that ecosystem, if you like, is a lot of aggregation. So you've got big companies, they tend to go to large registrars, and those large registrars are generally global in nature, and those registrars will pick up country codes if they're easy to add, and they'll add them if they can just literally do go to log into an interface and select that they'll now turn on.au and because they've got a lot of automated software underneath that, within seconds, they can actually add a country code to their to their service offerings, as long As it's actually with one of those bigger registrars.
So the challenge here in the Pacific region is that a lot of countries are small. They'll have less than 10,000 names, and so large registrars will sort of go, Well, we're happy to take on a new domain space if it's worth hundreds of 1000s of names or millions of customers, but they don't really want to put the effort into creating a separate business process to connect to a small country that might have different software requirements. You know, manual forms to fill out, like what Russell's saying, because their systems are all fully automated, that they won't support any country that's manual. It needs to be automated, but the option, and as Andrew says, the option to think about, is, could a group of Pacific ccTLDs in this region actually aggregate their operations? So you could have 20 country codes in the region, each of which is less than 10,000 names, but 20 collectively, that's of interest to registry operators that can provide software. If a registrar can pick up 20 names in the region through a single interface, that's of interest to the registrar, and then if the registrar has got those 20 names from the local region enabled, then if you're a company like Squarespace and you're already connected to that registrar, you just go sure, you know, we've got customers that would be interested in those names. If it's easy to do, they'll just turn them on. So one of the things that we're sort of considering is, is there ways that we can sort of aggregate that demand potentially select a registry operator in the region that has local support and basically provides a sort of reliable infrastructure. I've got a few slides. I'll go through these really quickly just to pick up on different things. But some of the other benefits is of aggregating together, is reliability. So if you're running a registry just in one country, and you have some natural disaster like a campaign or typhoon or an earthquake or fire, and it can take down all your infrastructure, you really want to be on a shared infrastructure where one node of that infrastructure, if it's damaged or unavailable, the services from that country are available at one of the other nodes. So it's sort of not just here about aggregating to get into a sort of a bigger infrastructure, but it's actually having connection into a much bigger shared infrastructure that's more resilient to climate change and natural disasters that we've talked about. And then the other aspect of it is that there's more and more cyber criminals that are exploiting business processes similar to email, obviously. So the big email providers, like Gmail and Microsoft, they invest a lot in trying to stop viruses and malware that can appear on their email systems. And the same thing with registry operators. They invest a lot to defend against cyber criminals and cyber attacks, and so it's increasingly hard to do that when you're small in scale. You really want to benefit from an organization that's spreading those costs over a large number of names, which comes back to sort of keeping your costs down, because if you were taking up the full cost infrastructure, multiple DNS servers spread across into multiple locations with multiple communication links, you build up a DNS abuse team. You're building up a cyber security team. If you're doing all that at the price of 10,000 names, then you'd be charging $1,000 a name. So, you know, just starts pricing, pushing the price up too much. So you've got to find ways of sharing the resource. I've got some slides, but given time, I will maybe just jump to bits and pieces of some of these. These slides will be available online. But one of the things on the DNS for resilience is each country should try and run their own DNS name server within their country, and that means that if that country was disconnected because the cables were broken or potentially attacked by some sort of attacker, or there was an earthquake and it ripped up the cables or something. You still want the people within your country to be able to access the DNS. You want a DNS name server there, but if everybody's got a D running each of you could run a DNS name server with, say, 100,000 names on it, and you'd have the names of all you call it, all your surrounding regions on each name server. So if any one name server goes down, the country names are still available on the other name servers. One of the groups that's actually done this is in Latin America, similar to the Pacific. There's lots of countries in the Caribbean region. And lac TLD has actually developed a service which is a sort of shared DNS infrastructure. And so any of the Caribbean countries can sort of use that shared infrastructure. And so that might be a model for considering in this region as well, that we sort of plan and build out a DNS infrastructure. Bits of it are in every country, but you've all got each other's ccTLDs on that infrastructure, and if any one node breaks, then the services are available from one of the other nodes. So this gives you that redundancy, but it also builds capability in your country. So by running your own DNS name server, you can train up your people and have a degree of expertise and DNS knowledge, but you're sharing a common infrastructure across the region. Just jumping various things here about complaints and deanus abuse, one of the other things, and I was conscious of the sessions this morning, but it's not just about infrastructure. So I think a few people said that in some countries, there's a lot of effort to put in a telecommunications cable, but not a lot of effort to sort of develop local economy to make use of that communications cable, and the need to sort of do that, put about as much effort into that than into the infrastructure. So one thing here is that if you can aggregate your infrastructure and have a high quality infrastructure for the collection or the region of Pacific ccTLDs. Then within each country, you can put effort into actually getting the local community to use that ccTLD and some of it really is, you know, basic, firstly, reminding consumers of what the country is. One of the things that we really promote in.au is that.au? Means you're Australian. So we have some checks that we do to make all the people that register names are Australian, so it gives it authenticity, but really reminding people that you know the.au at the end of the name means Australian, or the.ck at the end of the name means Cook Islands. The second part is basically building awareness of the benefits of of having your own domain name versus having an Instagram account or a Tiktok account or something not that you should this shouldn't be exclusive, because it's good to have a presence across a number of these platforms. But there's some benefits, and it's kind of a separate presentation, but they're definitely benefits of having domain names compared to some of these other alternatives, particularly in terms of in control of your own destiny. Because some of these platforms can turn you off overnight. You might have 100,000 followers, and for whatever reason, they can just close your account and you don't really have much come back. Whereas, if you own your own domain name and your own your own content, you're in control of your destiny. And then the third part of this is about, sort of demonstrating that it's easy. So you basically start with awareness raising. What's the country code for? What are the benefits compared to other online alternatives. And thirdly, just show people how to do it. Literally show someone that you can build a website in five minutes, and so that they then removes that barrier that they think that's there, that it's too hard. It's actually trying to make it, you know, help people realize it's not hard, and they can, she control their own sort of online presence. So that's anyway, a basis of an idea, and more than happy to sort of talk about it more during the week. And there's some slides that go into a bit more detail in those different areas. But the concept is essentially to say there's some benefits in aggregating resources across the region towards some sort of common benefits. So Linda,
thank you, Bruce. I'm sure if you have questions, please reach out to Bruce and a very interesting opportunity here that he has presented for ccTLDs, especially in the in the Pacific region. So if you're interested to talk more to Bruce, please do so. Our final presentation to this afternoon would be from chappeca. I'm not sure Champika is here. Oh yeah, he's here. Skipped him, so he's from ICANN. Many of you might know about ICANN. ICANN is a multi stakeholder engagement group, but it has a lot of acronyms. But he will explain what the acronyms mean. Thank you. Jim picker,
Thanks, Andrew. I know it's six past, but I'll try to be very quick, so I'll be actually talking about the universal acceptance. So this is a topic probably some of you are already familiar. So let's go into that straight now, when you think of DNS, of course, you know, we can think of different eras, right? DNS is about 40 plus years ago we started. And then, if you think of pre 2009 in fact, during early 80s or so, we still had these three letter generic, TLDs, the legacy ones.com.net, and so on. And then we also started having this country called top level domains. We already know that the ASCII ones. And then after 2009 or so, we started having these IDNs ID and countercode top level domains. And then after 2013 or so, we started having these new generic top level domains, and the ID and gTLDs as well. So these things, these domain names, have evolved over the time. So our focus is going to be on the IDNs, or internationalized domain names now. As of now, we have delegated 62 country code scripts for 43 countries in the world. Now, when we speak of these scripts, basically these are country code, Country Code CTL descripts. And for example, if you take country like in India, they they have several different language scripts. So you you'll see quite a lot of scripts. The IDN scripts have been delegated to India. And similar to, you know, there are different other scripts delegate to other countries as well. Now talking about the universal acceptance. The whole idea here is that all these domain names, earlier, we were talking about different categories of domain names. We had these legacy ones, and then we also had the new generic top level domains, the long ones, and then the IDNs so if you're having any applications, basically, you know all these domain names should work with all these applications. When I say domain names, it includes the email ids as well, because part of your email ID is also a domain name. There are few examples here. You can see that if you think of IDN, or even if you think of a new gTLD, or even if you think of a internationalized email address, all these things should work in your applications. Say, for example, if you register for this conference and if you have a internationalized email ID that you try to give when you register, and if the registration form says it's an invalid email id, that means there's a problem. So this is where we say that it is not universal acceptance ready? So the whole idea of universal acceptance is that make sure that all your software applications could support all these domain names and the email addresses that we are talking about now, when you think of the UA readiness or the universal acceptance readiness, there are few principles that we are talking about. First of all, acceptance. So if you give some input, it should accept it and then the validation process. It should, you know, make sure the validation happens. Then, if you're going to store these email IDs or domain names in a database, it should be, we should be able to store that and then process it and also display that. Say, for example, if it's a browser, it should display it properly, right? So these are the important principles that we have to think from a UA perspective. Now these are, again, you know, some examples for you to understand this clearly. So if you're going to type in your browser, say the first one.org, which is, you know, just a new gTLD. So browser should display that. If it's not displaying it, that means there's a problem with that application. So we say that UA not ready. I mean, it's not ready, right? And then, similarly, the second example is that if you're going to have IDN, TLD over there, so if the browser can display that, then that's UA red. But if it can't, yeah, you know, we have to fix that. And similarly, if you see the third one, you'll see in your there are some email IDs that we try to display that's in local language scripts. So if it's not going to display that properly, yeah, there's a problem. So you are ready versus not ready right now, when it comes to the email ids, we say that email address internationalization, or simply we call it as EAI. So basically, your email ID can consist of, you know, different categories, different formats. So when we say EAI, so we are talking about having UTF eight support, right? So now, the thing is that, you know, when you go back to historical days, we use ASCII encoding now, ASCII, with ASCII, we only have limited sort of characters that character sets we could use, say, for example, in a domain name we know that we could use say, if it's a TLD, we could use from A to Z, and then, if it's a second level domain, we can use from A to Z and zero to nine, and then the hyphen. But if it is anything more than that, say, a local language, character in a given language, we can't use ASCII because ASCII table is also full, so we have to go into some other encoding scheme. So this is where we go for Unicode based encoding schemes. There are several encoding schemes in Unicode. But we go for this UTF eight, which is Unicode transformation format, eight bits, basically, this is the most optimized format for these applications that we are talking about. Because one thing is the backward compatibility with ASCII, because you could still work with ASCII as well, and the other thing is the storage optimization and transmission. So UTF This is why UTF eight has become a standard now. So when you have UTF eight support, that means when you think of an email ID, so you have two parts in your email ID, you have the mailbox part, and then you have the domain part. So if you can.
