Latest / Crypto Altruists: Real-World Stories of Social & Environmental Impact with Web3 / Episode 270 - AI and Blockchain for NGOs: How Decentralized Technology Is Driving Innovation in International Development, with FLock.io

Transcript

Episode 270 - AI and Blockchain for NGOs: How Decentralized Technology Is Driving Innovation in International Development, with FLock.io · Crypto Altruists: Real-World Stories of Social & Environmental Impact with Web3

• 0:00
  SPEAKER_01: Today you'll discover how community-based or decentralized AI and blockchain can help NGOs and nonprofits innovate without handing over their data and power to big tech.
• 0:09
  SPEAKER_01: Look into Flock.io's work as the AI strategic partner for the UNDP's SDG Blockchain Accelerator and the projects transforming lives across Africa, Asia, and Latin America.
• 0:18
  SPEAKER_01: How this technology is getting aid to people faster and helping communities capture the value they create rather than losing it to intermediaries, and how the Blockchain for Good Alliance is bringing together the UN, universities, and grassroots builders to turn good ideas into real-world impact.
• 0:34
  SPEAKER_01: Welcome to CryptoAltrux, where we explore the intersections of Web3 and Social Good to the stories of impact-driven builders and offer actual insights to empower your journey into the world of blockchain for good.
• 0:42
  SPEAKER_01: I'm your host, Drew Simon.
• 0:44
  SPEAKER_01: Now before we dive in, please remember that this is not financial or legal advice, and always do your own research on any of the projects or investment opportunities that we discuss.
• 0:52
  SPEAKER_01: Welcome back to the Crypto Alterist Podcast.
• 0:54
  SPEAKER_01: Thank you so much for joining us today for episode 270.
• 0:57
  SPEAKER_01: If you work in the nonprofit or international development space, you've probably felt the pressure lately.
• 1:01
  SPEAKER_01: Funding is getting tighter every day, it seems.
• 1:04
  SPEAKER_01: Needs are growing faster every day, it seems, and everywhere you look, someone is telling you that AI will change everything.
• 1:09
  SPEAKER_01: But for many NGOs and civil society organizations, the real question is much more practical.
• 1:14
  SPEAKER_01: It's where do we start and how do we do it without putting the communities we serve at risk?
• 1:19
  SPEAKER_01: In episode 269, where we did our solo episode on sovereign AI, we explored how some organizations are using AI in ways that keep data in the hands of communities.
• 1:27
  SPEAKER_01: One of the examples we highlighted was the work of Flock.io with the United Nations Development Program, and today I'm thrilled to go much deeper on that story.
• 1:34
  SPEAKER_01: I'm joined by Tiffany Wang, head of EMEA at Flock.io, and a team member of the Blockchain for Good Alliance.
• 1:40
  SPEAKER_01: Flock is building decentralized AI, an approach where models can learn from data without that data ever leaving communities it belongs to.
• 1:47
  SPEAKER_01: It's a powerful alternative to the dominant model where AI is trained by a few large companies on data gathered from everyone else.
• 1:53
  SPEAKER_01: And they're putting it into practice at a global scale as the AI strategic partner for the UNDP's SDG Blockchain Accelerator, supporting projects that span the developing world.
• 2:01
  SPEAKER_01: We're talking microinsurance for smallholder farmers in the Dominican Republic, artisans in Morocco, healthcare in Mauritius, and faster aid delivery in Liberia.
• 2:08
  SPEAKER_01: There's some really cool stuff going on.
• 2:10
  SPEAKER_01: It's really incredible seeing the diversity of projects involved, and they really show what becomes possible when AI and blockchain are designed around the needs of last mile communities.
• 2:18
  SPEAKER_01: Tiffany also works with the Blockchain for Good Alliance, a collaborative hub bringing together global organizations, universities, and builders to strengthen the use of blockchain for social impact.
• 2:27
  SPEAKER_01: Today we'll talk about data sovereignty and why it matters so much for vulnerable communities, the pilots coming out of the UNDP accelerator, how this technology can help communities cut out the intermediaries that have historically captured the value they create, and what it will take for decentralized community-owned AI to become the norm rather than the exception.
• 2:43
  SPEAKER_01: And for any NGO or civil society organization wondering where to start, we'll get into that too.
• 2:48
  SPEAKER_01: So let's dive in.
• 2:51
  SPEAKER_01: Tiffany, thank you so much for being here today.
• 2:53
  SPEAKER_01: I am so excited to dive into this fascinating conversation at the intersection of AI, blockchain for good, and some really cool work they're doing with communities, including the UNDP, really all over the world, to support some fascinating projects that are using tech for some really cool novel uh innovations.
• 3:10
  SPEAKER_01: So thank you so much for being here.
• 3:12
  SPEAKER_01: I'm really excited for this conversation.
• 3:14
  SPEAKER_00: Thank you so much for having me.
• 3:16
  SPEAKER_00: Um, and we're looking forward to the conversation as well.
• 3:19
  SPEAKER_01: Yeah, yeah, me too, me too.
• 3:20
  SPEAKER_01: Great.
• 3:21
  SPEAKER_01: Um, you know, I'd love to learn about you to kick things off.
• 3:23
  SPEAKER_01: Uh, you have a really interesting uh kind of background and journey to Web3 uh and AI, and you're kind of here representing both uh Flock.io and uh which is kind of like an AI, decentralized AI kind of uh startup, uh, and then Blockchain for Good Alliance, which many in my community might be aware of that does a lot of great work in terms of like elevating uh Web3 impact use cases.
• 3:45
  SPEAKER_01: So maybe just like tell me a little bit about your journey and how you got involved with uh Flock.io and Blockchain for Good Alliance.
• 3:52
  SPEAKER_00: Mm-hmm.
• 3:53
  SPEAKER_00: Um thank you for asking this question, actually.
• 3:55
  SPEAKER_00: Um I am not a crypto crypto native or AI native.
• 3:59
  SPEAKER_00: Uh I come from legal background.
• 4:01
  SPEAKER_00: So uh my role of blocking, like a break into AI and blockchain industry was, I would say, um, a little bit random, but I I've always wanted to, you know, do something to create impact.
• 4:15
  SPEAKER_00: It sounds a bit cliche, but it actually has been my childhood dream to work for United Nations.
• 4:20
  SPEAKER_00: And turns out we're, you know, like we're working with them as their you know strategic AI partner, blockchain partner.
• 4:26
  SPEAKER_00: So, you know, uh I think it's just like one thing leads to another.
• 4:30
  SPEAKER_00: I joined as a legal counsel at Flock First, and then I made my way up, and then I because uh because of my my role and responsibility get increased over time.
• 4:41
  SPEAKER_00: So I'm also doing like um business part of stuff, and then through that I um I happen to know the blockchain for good alliance and I admire what they do, and then now I'm also helping the blockchain for good alliance as well.
• 4:54
  SPEAKER_00: So I guess it's like um it's it's more like an a natural uh evolution, if that makes sense.
• 5:00
  SPEAKER_00: Like it's because I'm interested in what they do and want to understand more, and happen to be like that a lot of the things that I do are intertwined with each other and actually mutually beneficial as well.
• 5:11
  SPEAKER_01: Yeah, yeah, that's really cool.
• 5:12
  SPEAKER_01: Uh well I'm excited to dive into that and and what you're working on right now.
• 5:15
  SPEAKER_01: And so maybe we'll start with um the work that you're doing with Flock.
• 5:19
  SPEAKER_01: Uh so can you break it down in plain language?
• 5:21
  SPEAKER_01: Um, you know, what Flock does and the intersection of AI and blockchain and how this plays into it.
• 5:28
  SPEAKER_00: Yeah.
• 5:29
  SPEAKER_00: So um we I don't want to talk about what is decentralized AI, what is federal learning, right?
• 5:35
  SPEAKER_00: Let's just say like um everybody probably, you know, to a certain extent, you're using some sort of AI tools, right?
• 5:42
  SPEAKER_00: And then for all those AI tools that you are seeing in the market right now, it's it you can think about it as like a centralized AI platform.
• 5:49
  SPEAKER_00: What we are building is a decentralized AI infrastructure where users do not have to transfer or use OSEO data to a centralized platform.
• 5:58
  SPEAKER_00: Essentially, you don't contribute your own private data for us for model training.
• 6:03
  SPEAKER_00: The model we use Federal Learning as a training method to help you help a community to train, co-train a model, collaborative training, uh, but without anybody sharing their raw data.
• 6:16
  SPEAKER_00: It's only the learning of that data.
• 6:18
  SPEAKER_00: I can give you maybe an example, right?
• 6:20
  SPEAKER_00: Like you can be a group of, you know, like doctors or a um, I don't know, scientists talking about the learning of a certain type of disease, the new discovery, but without exposing a patient information that makes sense, right?
• 6:33
  SPEAKER_00: You only talk about the learning, and uh, and that's essentially what federal learning does.
• 6:37
  SPEAKER_00: And Flock is not only just federal learning, right?
• 6:40
  SPEAKER_00: Because federal learning is actually a technology that has been proposed by Google, it has already been used.
• 6:46
  SPEAKER_00: What we have done is like we add a blockchain verification layer on top of the federal learning training round.
• 6:52
  SPEAKER_00: Why does it why does it matter?
• 6:54
  SPEAKER_00: Because in normal federal learning uh environment, you don't control the node.
• 6:59
  SPEAKER_00: Like we call it training node.
• 7:01
  SPEAKER_00: You you say, okay, okay, everybody can join the training round, but you don't really know what kind of contribution that each individual individual has made.
• 7:09
  SPEAKER_00: You don't you have no way to verify, no way to trace, and no way to get either penalized or you know give people reward.
• 7:17
  SPEAKER_00: There is the the part where we add the blockchain verification layer is essentially what it why why uh how uh how these things can happen.
• 7:24
  SPEAKER_00: Because blockchain is a ledger, right?
• 7:26
  SPEAKER_00: You record everything, and at the same time, we added a small contract uh mechanism to calculate and the reward and a contribution for that model.
• 7:34
  SPEAKER_00: Yeah.
• 7:35
  SPEAKER_01: Yeah.
• 7:35
  SPEAKER_01: Yeah, that's really cool.
• 7:37
  SPEAKER_01: And you mentioned um, you know, like not having to give away your data to to big like tech and and and those players and the more centralized models.
• 7:45
  SPEAKER_01: And I want to zero in on this concept of data sovereignty because I think it's increasingly important in our world.
• 7:50
  SPEAKER_01: Uh, the idea that communities and individuals shouldn't have to hand over their private data to be able to benefit from AI technologies.
• 7:58
  SPEAKER_01: And so why is this so important, especially when it comes to like vulnerable and last mile communities?
• 8:03
  SPEAKER_00: Yeah.
• 8:04
  SPEAKER_00: Um I I wouldn't, okay, let's let's not be like dramatic, right?
• 8:11
  SPEAKER_00: I wouldn't say like, oh, all the information that you have is so critical and important that you cannot use any, you know, general AI tools.
• 8:19
  SPEAKER_00: You know, like don't get me wrong, like I also use uh I also use Clot, but I wouldn't trust it for everything.
• 8:26
  SPEAKER_00: I wouldn't tell him everything that I do, right?
• 8:29
  SPEAKER_00: Because some kind of information, for example, your healthcare data, for example, I don't know, um your relationship information with certain type of people, your your opinion for some type of people, those kind of like sensitive personal information.
• 8:43
  SPEAKER_00: And why to answer your question, why this is so important for, you know, um and bank groups of people who are vulnerable, because the reason why they're vulnerable is because either, you know, they are not accepted by mainstream, uh, mainstream society, maybe because of their religion, maybe because of their you know, belief, maybe because of their sexuality, all those kind of reasons make them vulnerable.
• 9:04
  SPEAKER_00: And if you uh and that's essentially why we think for those important vulnerable groups, we you we need to add an actual privacy privacy or as actual sovereignty to protect them when they are using AI tools.
• 9:18
  SPEAKER_01: Yeah, no, I think that that's that's super important.
• 9:20
  SPEAKER_01: And I love that you're doing that.
• 9:21
  SPEAKER_01: We're gonna talk about examples of kind of how that's happening with some of the work that you're doing.
• 9:25
  SPEAKER_01: Um, but I'd love to, you'd mentioned in the kind of when we were introducing each other that uh it's been your dream since you were a child to work uh for the UN or with the UN.
• 9:34
  SPEAKER_01: And it's really cool how that's kind of happening now.
• 9:36
  SPEAKER_01: Um and you know, I know that Flock has been working with the United Nations Development Program uh as an AI strategic partner uh for their SDG blockchain accelerator, which is uh super cool.
• 9:47
  SPEAKER_01: Um can you tell me more about this and how this uh kind of all came together and what the accelerator is and how it works?
• 9:53
  SPEAKER_00: Yeah, so um I'm not sure if if you know what UNDP does.
• 10:00
  SPEAKER_00: Like UNDP inst UNDP is an agency within United Nations, right?
• 10:04
  SPEAKER_00: And you focus on developing countries.
• 10:06
  SPEAKER_00: And a lot of people might have this, you know, misconception or like even misunderstanding towards, you know, um UNDP is acting slow, you know, they are you know bureaucratic, you know.
• 10:18
  SPEAKER_00: This is what I assume as well.
• 10:19
  SPEAKER_00: And but once I started working with them, I was like, there's actually an innovation team.
• 10:24
  SPEAKER_00: You know, they they are quite curious and quite uh advanced in terms of technology discovery.
• 10:30
  SPEAKER_00: They I they're actively working with the private sector.
• 10:33
  SPEAKER_00: So that's how the whole idea of accelerator comes from, right?
• 10:37
  SPEAKER_00: We were doing pilots, we're doing accelerator cohort one, two, three.
• 10:41
  SPEAKER_00: So with cohort one, we have you know like two projects, like I think it was like a like more like a testing phase.
• 10:47
  SPEAKER_00: And then with cohort two, we expanded the scope, you know, to 50, I think it's 50 countries, develop only developing countries, and each country will propose a count uh a specific problem that they need to be addressed by blockchain AI solutions.
• 11:03
  SPEAKER_00: And where we come in as technical AI partner will help them to find solution maker, or I will help to find them industry, you know, um in industry leaders, and then we we help them to design the solution employment, and sorry, and and and deploy that solution in the local context.
• 11:20
  SPEAKER_00: And that's how the whole idea comes from, because we don't want to just say, okay, um with UNDP, we're going to do, you know, like uh another pilot.
• 11:29
  SPEAKER_00: We want to actually have the pilot be implemented.
• 11:32
  SPEAKER_00: So um so yeah, it's you can think about it as like accelerator is is um it's a it's a trial, is it even within UNDP, and it will keep in working and refined the method of how do we push the last milework to finish, how do we make the solution actually implementable and create the impact in local context.
• 11:54
  SPEAKER_00: Yeah.
• 11:54
  SPEAKER_01: Yeah, and I love that it's like really community focused, community-driven, uh local, which is great.
• 12:00
  SPEAKER_01: Um, and you mentioned something at the beginning, which is interesting, around like, you know, the UN being viewed, and I think I had some of this view as well, maybe before I kind of got into this field of work around like, you know, the UN being all bureaucratic and slow and they're not really doing anything and those sorts of things.
• 12:14
  SPEAKER_01: But it's really interesting, and I released an episode on this uh recently, actually, that I very much view the UN as like a leader in the blockchain AI space for like for good, for NGOs.
• 12:24
  SPEAKER_01: Um they've been there since the beginning with their like crypto for good fun, with the building blocks program to the world food program.
• 12:30
  SPEAKER_01: Uh the UNDP now has their blockchain advisory committee.
• 12:32
  SPEAKER_01: Like, there's so much cool stuff happening like um with the UN.
• 12:36
  SPEAKER_01: They're really taking a leadership role in finding ways that this tech can be used for good.
• 12:40
  SPEAKER_01: And so my perspective has has really shifted on this.
• 12:42
  SPEAKER_01: And um, you know, how is it like have you found the same?
• 12:46
  SPEAKER_01: Like you you'd mentioned earlier on that uh there's a lot of great, you know, emphasis being put on blockchain and AI within the UN.
• 12:52
  SPEAKER_01: What sort of reception are you getting from the folks that are in the UN towards this kind of technology?
• 13:00
  SPEAKER_00: Um I I would say like you don't know what you don't know, right?
• 13:05
  SPEAKER_00: Uh we we we are the private private sector partner.
• 13:08
  SPEAKER_00: Like we like before working with UNDP, we think, oh, like the solution is already there.
• 13:16
  SPEAKER_00: Like we have designed the perfect solution.
• 13:18
  SPEAKER_00: Why can't you push it to finish?
• 13:20
  SPEAKER_00: Why can't you implement it?
• 13:22
  SPEAKER_00: And it's not that easy.
• 13:23
  SPEAKER_00: It is really not that easy, right?
• 13:26
  SPEAKER_00: Because when it when it comes to you know driving a big change, UNDP has to go through the internal, you know, approval, and they have to make sure that this this solution is actually safe, secured.
• 13:39
  SPEAKER_00: And that's why when you talk about you know those kind of private sector groups, like we as FOLC also being selected as a member of the blockchain advisory group.
• 13:48
  SPEAKER_00: It's not like, okay, um, so we think this solution will work, and then then UNDB is gonna implement it immediately.
• 13:55
  SPEAKER_00: It actually has to reach a conclusion, and then um UNDB has to serve as a bridge between the private sector and the local government, and they have to help us find, okay, who is the right person to talk to.
• 14:07
  SPEAKER_00: Do you actually understand, for example, the local contact Dominican Republic?
• 14:11
  SPEAKER_00: You think you are you think this solution is perfect for the Dominican Republic.
• 14:15
  SPEAKER_00: But if you don't, if you don't know, for example, like we also have a solution in Liberia, right?
• 14:20
  SPEAKER_00: And we were we were being educated saying, like, okay, this solution, originally the the dualist solution that we designed does not include feature phone and coupons because we assume that everybody has a mobile phone.
• 14:33
  SPEAKER_00: But that's not the case, right?
• 14:35
  SPEAKER_00: A lot of assumptions, a lot of a lot of like um, and I say, like you just take it for granted and you design a solution that you think is gonna work without taking in consideration of the local contact.
• 14:46
  SPEAKER_00: That's why, that's why I said like you don't know what you don't know.
• 14:50
  SPEAKER_00: So I think a lot of initiatives like UNDP or your other UN agencies, for example, for program, we're also working with them as well.
• 14:57
  SPEAKER_00: Very, very impressive.
• 14:58
  SPEAKER_00: Not just because um um they have you know made this grand gesture, but at the same time, they're the people who actually get to talk with local people, understand the real cause, and understand how we can actually help them to deliver the last mile work and create tangible impact.
• 15:15
  SPEAKER_00: It's not like just private solutions saying we complain, you know, like, oh, um, why this solution is not getting implemented?
• 15:21
  SPEAKER_00: Like, is it because you know, the the bureaucratic uh in the process or something?
• 15:26
  SPEAKER_00: Sometimes it's more complicated than that.
• 15:28
  SPEAKER_00: Wait, we need to think about a way the better we find our solutions.
• 15:32
  SPEAKER_01: Yeah, absolutely.
• 15:33
  SPEAKER_01: Um yeah, no, couldn't agree more.
• 15:35
  SPEAKER_01: And I'm excited to jump into some of the examples now of this work actually happening on the ground.
• 15:40
  SPEAKER_01: And you'd mentioned uh the Dominican Republic, you mentioned Liberia.
• 15:44
  SPEAKER_01: Um, I know that there's like some really cool projects taking part in this accelerator.
• 15:48
  SPEAKER_01: Microinsurance for smallholder farmers in Dominican Republic, guerrilla conservation in Rwanda, artisans in Morocco, healthcare in Mauritius.
• 15:55
  SPEAKER_01: Like there's so many cool, diverse projects really happening all over the world.
• 15:58
  SPEAKER_01: Are there any that like really stand out that you'd like to share more about um to kind of really show the value that this accelerator is bringing to these communities?
• 16:06
  SPEAKER_00: Yeah.
• 16:07
  SPEAKER_00: Um I would, I would, I wouldn't want to brag and say like, okay, all this initiative have you know delivered like um tangible impact immediately now.
• 16:18
  SPEAKER_00: I'll I was I would meet that a lot of projects we're doing are still in the process.
• 16:23
  SPEAKER_00: We didn't give up, you know.
• 16:25
  SPEAKER_00: It's hard, it's hard.
• 16:26
  SPEAKER_00: We have we we you know, through Accelerator, we you know, we coordinate with local stakeholders.
• 16:31
  SPEAKER_00: UNDB has been really helpful.
• 16:33
  SPEAKER_00: And then we designed the solution, we refined it.
• 16:36
  SPEAKER_00: Like I mentioned, you know, the Libera Digital Wallet solution, we refined it.
• 16:39
  SPEAKER_00: And then we we with their with their help, we started looking for local partners, you know, either, you know, Libera, we need to work with um, you know, telecom companies in Dominican Republic because a micro insurance product, we also need to work with local insurers.
• 16:54
  SPEAKER_00: So and at the same time, we need to uh navigate a regulatory industry, uh regulatory environment as well.
• 17:01
  SPEAKER_00: So um, and then that solution right now, we are um we have reached, for example, the Dominican Republic when we reached a milestone.
• 17:10
  SPEAKER_00: We signed an official framework agreement with the UNDP uh Dominican Republic, meaning that you know this this solution that we're building has been rec recognized by the local uh government, by official local government.
• 17:24
  SPEAKER_00: This has been put into their um uh I think it's like um strategic plan for the next five years.
• 17:30
  SPEAKER_00: Don't quote me on that.
• 17:32
  SPEAKER_00: It may be 10 years.
• 17:33
  SPEAKER_00: I'm not sure.
• 17:34
  SPEAKER_00: I'm not so sure.
• 17:35
  SPEAKER_00: Yeah, yeah.
• 17:35
  SPEAKER_00: Five to ten years, we'll say five to ten years, right?
• 17:38
  SPEAKER_00: So it's like an official, official commitment that we're working on this.
• 17:44
  SPEAKER_00: Um and the Liberia one, I hope soon it will also reach that stage as well.
• 17:49
  SPEAKER_00: So we're working on this.
• 17:50
  SPEAKER_00: We we we wanted we really wanted to help this um push this solution uh to you know, not only being implemented in the Dominican Republic, but it's you know, essentially that's the climate uh product, right?
• 18:02
  SPEAKER_00: Climate resilient product for any regions, any countries that has this kind of you know, um change like um how to say drastic climate disaster situations.
• 18:13
  SPEAKER_00: And um this solution can be replicated and applied in a different region if we can find a local partner to work with as well.
• 18:20
  SPEAKER_00: So I think this is really valuable and we wanted to expand it further.
• 18:25
  SPEAKER_00: Yeah.
• 18:26
  SPEAKER_01: Yeah, yeah, that's awesome.
• 18:27
  SPEAKER_01: Um I love that.
• 18:28
  SPEAKER_01: That's that's so cool.
• 18:29
  SPEAKER_01: Congrats on getting that agreement uh in place in the Dominican.
• 18:32
  SPEAKER_01: That's great.
• 18:33
  SPEAKER_01: Hopefully that leads to some really cool and um you know, you mentioned Liberia as well, and I want to kind of zero in on that too, because the pattern that I noticed when I was doing my research uh for this interview is that there is um across several of the pilots uh a focus and emphasis on kind of like the speed of relief, you know, getting money to people fast.
• 18:49
  SPEAKER_01: Um, you know, in Liberia, for example, blockchain payments uh are really cutting disbursement times from up to a week down to like almost real time, right?
• 18:58
  SPEAKER_01: Um with options for feature phones, paper vouchers for those without phones, so making it really accessible in addition to you know increasing that speed.
• 19:06
  SPEAKER_01: Um talk to me about this and why that is such an important focus and kind of some of the results you're seeing there.
• 19:12
  SPEAKER_00: Yeah.
• 19:12
  SPEAKER_00: Um I think you're talking about two fronts, right?
• 19:16
  SPEAKER_00: One is speed, and then it's accessibility, right?
• 19:19
  SPEAKER_00: Let's talk about speed first.
• 19:21
  SPEAKER_00: Why is it so important to make sure it's like the money that you need to get from um either your insurance product or is this like um um disaster release money, uh reimbursement or compensation?
• 19:36
  SPEAKER_00: Because the the whole thing is about protecting, like the safety net, right?
• 19:42
  SPEAKER_00: It's like you really need that money.
• 19:44
  SPEAKER_00: And if we don't have the money, maybe you have to, I don't know, in in those cases that we have heard, the family they're relying on that for food or you know, kids' education, or uh sometimes they, you know, they they have don't have that money uh immediately now, they have to sell their house.
• 20:02
  SPEAKER_00: Or all those kind of like urgent situations.
• 20:05
  SPEAKER_00: That's why when when the disaster happens and then the the money that you know the the money that needs to come in immediately, otherwise people lose trust in it, and uh and they will get people less incentive to to vote in that solution in the first place.
• 20:20
  SPEAKER_00: Because going back to our point, like the solution that we are building are targeting at vulnerable groups.
• 20:26
  SPEAKER_00: And maybe that, you know, I don't I don't I I'll take, for example, maybe that hundred dollar for you, it does not matter.
• 20:32
  SPEAKER_00: It can come in today, it can come in tomorrow.
• 20:34
  SPEAKER_00: But if that$100 is like uh is at your your family's food, then you want you want it to come in immediately, isn't it, right?
• 20:42
  SPEAKER_00: Because it's so important for you.
• 20:43
  SPEAKER_00: And that's the speed part, why we think it it we need to get we need to get this money to people's hand as fast as possible.
• 20:51
  SPEAKER_00: And why accessibility is important?
• 20:53
  SPEAKER_00: Because that's a whole value why we even need to design the solution in the first place.
• 20:58
  SPEAKER_00: Because we want to include everybody.
• 21:00
  SPEAKER_00: If we say, okay, all the solutions that we're getting, you know, uh all the AI products or everything that we design for people who own people who live in the first um first-tier cities, people who who um who not inoper from you know London or New York, and that's that's not a real impact.
• 21:18
  SPEAKER_00: Because then that's what, like one percent of the society, you're leaving 99% of people behind.
• 21:26
  SPEAKER_00: That's not the value of why UNDP needs to do this accelerator.
• 21:30
  SPEAKER_00: That's not why, you know, we are even focusing on regions like Dominican Republic and Liberia, where we don't design a solution in London, right?
• 21:38
  SPEAKER_00: Because there are tons of solutions already.
• 21:40
  SPEAKER_00: And then where people don't need this specific inclusive element of that solution.
• 21:46
  SPEAKER_01: Yeah, no, absolutely.
• 21:47
  SPEAKER_01: And I think that um it's great that it's you know really focusing on building for communities kind of in like meeting them where they're at, as opposed to trying to like, you know, impose like uh technology on communities and saying, hey, you need to use all these new tools.
• 22:00
  SPEAKER_01: Well actually you're using this already, you have a feature phone, uh, you have paper vouchers that you use, you know how to do that, so we'll work with that, and I think that's a great way to kind of meet people where they're at.
• 22:09
  SPEAKER_01: And we'll be right back after this brief announcement.
• 22:12
  SPEAKER_01: Here's an interesting statistic.
• 22:14
  SPEAKER_01: More than 9 in 10 nonprofits are already using AI, but most of them have no policy for it.
• 22:19
  SPEAKER_01: If that sounds like your organization, we can help.
• 22:21
  SPEAKER_01: CryptoUltras offers practical consulting for nonprofits navigating Web3 and AI.
• 22:25
  SPEAKER_01: That might mean writing a responsible AI policy your team will actually use, finding real-time savings in your day-to-day work, setting up crypto donations, or using blockchain tools to build trust without putting your community's data at risk.
• 22:36
  SPEAKER_01: There's no hype, there's no jargon, and we'll tell you honestly when a tool isn't worth your time or this technology just isn't a good fit.
• 22:42
  SPEAKER_01: Every engagement starts with a free conversation, so to learn more, head to cryptoaltrist.com slash consulting, or simply click the link in the show notes.
• 22:49
  SPEAKER_01: And now, back to the episode.
• 22:51
  SPEAKER_01: You know, speaking of um kind of communities and and um you know, meeting people where they're at and that kind of localization element, I want to talk about value creation as well.
• 23:00
  SPEAKER_01: You know, getting into kind of crypto and web three, for me, a big thing we've talked about is like shifting value away from centralized actors or you know, big tech or whoever it is to communities and giving them more kind of sovereignty and power over things.
• 23:13
  SPEAKER_01: And, you know, another recurring theme I saw that as I was going through this work and researching is um, you know, fairness and benefit sharing being a big part of it, making sure that when communities create value, they capture it as opposed to you know giving it to some sort of intermediary.
• 23:28
  SPEAKER_01: Um, how do you see this technology helping uh nonprofits, communities, uh cut out the intermediaries and gatekeepers that have historically excluded these communities and giving them more power?
• 23:39
  SPEAKER_00: I would say um, I mean, if you talk about like how do we empower communities, especially the vulnerable groups, I would say like the solutions, when we design the solutions, we design it as design it to be as inclusive as possible so that they can also enjoy the benefit of technology, they can receive their payment as soon as possible.
• 24:01
  SPEAKER_00: That's the first part.
• 24:02
  SPEAKER_00: Secondly, fundamentally, what we believe at Flock is we think when when people contribute to that um can that model training, and we, you know, uh either you contribute by your data or your learning of the data or your um or your local compute, as in like you join the training round as a trainer.
• 24:22
  SPEAKER_00: We think every contribution you make, you need to be rewarded instead of saying like, oh, um I'm helping uh when I'm using the AI tools, I'm helping the big tech to train their model.
• 24:34
  SPEAKER_00: At the same time, I pay tokens, right?
• 24:36
  SPEAKER_00: Sorry, at the same time I pay tons of credits, and the more that I use, the model gets better, and the more the more money they charge me.
• 24:43
  SPEAKER_00: Like it does not feel right, isn't it?
• 24:45
  SPEAKER_00: Like I'm helping you to train your models at the same time, you earn all the money from me, right?
• 24:50
  SPEAKER_00: That's so we believe that uh not to a model not your model, not your AI, like if you help with training the model, you should own a piece of that model.
• 24:59
  SPEAKER_00: That's just how uh we we that a fundamental belief that we have, yeah.
• 25:04
  SPEAKER_01: Yeah, and I love that.
• 25:05
  SPEAKER_01: And I think that you know, that's really what I'm trying to like uh you know push towards with the work that I'm doing.
• 25:11
  SPEAKER_01: And you know, I think right now there is a lot of anxiety, right, around big tech and these big companies kind of concentrating the power and wealth when it comes to AI, right?
• 25:19
  SPEAKER_01: And I love that your model kind of looks the other way, really looks to make it community-owned, decentralized, kind of give them power for that.
• 25:27
  SPEAKER_01: How do we get to a point where that's kind of the norm rather than the exception?
• 25:32
  SPEAKER_00: Um I think in um let me let me think how to how to explain this better.
• 25:42
  SPEAKER_00: First, we need to start with education, um, because not many people know there is an alternative way of AI, like, you know, because mainstream AI companies, they're so big and they their performance really well, right?
• 25:57
  SPEAKER_00: Because um, like it's it's like no-brainer.
• 26:01
  SPEAKER_00: Like everybody's using ChatGPT, everybody's using anthropic, like, you know, all these kind of AI tools.
• 26:07
  SPEAKER_00: That's why people start using it more.
• 26:08
  SPEAKER_00: That's why the model gets better.
• 26:10
  SPEAKER_00: It's because the community contribute to it.
• 26:13
  SPEAKER_00: I think when to answer your question, how do we make sure, like, you know, um the decentralized AI model, decentralized AI infrastructure gets better, is also relying on community, right?
• 26:24
  SPEAKER_00: The more, the more people could build on it, the more people believe on it, the more people talk about it, the more people try to contribute on the whole whole uh decentralized AI infrastructure, the better it gets, right?
• 26:36
  SPEAKER_00: And the fundamental difference is not like how um how the it's not be it's not like um how the model performs.
• 26:44
  SPEAKER_00: It's like the the centralized AI, the centralized AI platform believe that the power is but the power belongs to the the the you know the AI companies who own the model.
• 26:56
  SPEAKER_00: And what we have done is we give the model power model ownership back to people.
• 27:01
  SPEAKER_00: But still we rely on people.
• 27:03
  SPEAKER_00: We need the people, we need people to believe in it.
• 27:06
  SPEAKER_00: We need people that oh, the ownership actually belongs to me, like I own the model because I I helped train the model.
• 27:14
  SPEAKER_01: Yeah, no, for sure.
• 27:15
  SPEAKER_01: And it's an interesting kind of mindset shift um that I'm I think a lot of people are still trying to wrap their heads around.
• 27:20
  SPEAKER_01: And there's been so much happening in the AI world, uh a lot of updates, a lot of new models being launched, uh just a lot.
• 27:26
  SPEAKER_01: And um, you know, what are you what is your take on kind of like where things are at?
• 27:30
  SPEAKER_01: A world of AI and the tools that you're offering for Flock.
• 27:33
  SPEAKER_01: Um, I'd love to kind of hear, you know, what you've been working on, what's in the pipeline for Flock.
• 27:37
  SPEAKER_00: Yeah.
• 27:38
  SPEAKER_00: So um, as you said, like, you know, the AI world is is updating, you know, not in a monthly base, but like daily base, right?
• 27:46
  SPEAKER_00: There are always new things happening.
• 27:48
  SPEAKER_00: Even me who work very closely in the industry in the center of the AI bus world, I also have the fear of missing out.
• 27:55
  SPEAKER_00: But I think one thing that we do need to understand and pay more attention is the cost of the mayor models of of you know how much you use the credit for and what kind of benefit you're actually enjoying from it.
• 28:06
  SPEAKER_00: So um, because we believe in decentralized AI, we believe in open source AI, and we believe that you know everybody should have the sort, should have the ownership of AI, should have the power to use it, benefit from it.
• 28:19
  SPEAKER_00: So we recently launched a model.
• 28:21
  SPEAKER_00: Uh it's a it's a bit similar to JAF model, it's a classifier model, and you can use it to make very fast decisions and the accuracy rate.
• 28:30
  SPEAKER_00: We've also done a benchmark right now, and it's uh we have the accuracy rate higher than other um open open market uh model as well.
• 28:38
  SPEAKER_00: And essentially, because we open source that model, so everybody can now just enjoy it for free.
• 28:43
  SPEAKER_00: And because it's a classifier model, it's not a reasoning model, you cannot expect it to you know make a complex complicated analysis, but they can use it for very simple multi-choice, or you know, just what's up, or or maybe like yes or no questions, and it comes back the answer to you in milliseconds, right?
• 29:02
  SPEAKER_00: So it's super fast.
• 29:03
  SPEAKER_00: And the model can be deployed in your local edge devices, it's very, very small.
• 29:08
  SPEAKER_00: You can even deploy on your phone as well.
• 29:10
  SPEAKER_00: So I think it's it's it's gonna it's gonna bring a lot of like uh benefits for normal people, you know, ordinary people who do not have the monetary power to buy large GPU, but they all can also train and use a model on your edge devices.
• 29:25
  SPEAKER_00: Yeah.
• 29:26
  SPEAKER_00: We're very excited for this new model.
• 29:28
  SPEAKER_01: Yeah, and it's it's interesting because I know that in a lot of you know countries, uh, you know, developing countries or you know, last mile communities, they might not have, you know, like you said, that GPU power or those like new phones with the you know, updated everything that can run these bigger models optimally.
• 29:45
  SPEAKER_01: They need like those smaller, lightweight models.
• 29:47
  SPEAKER_01: And so having something that you can like run on your phone or on like a small kind of computer is like really is a game changer when it comes to like sovereignty if you know organizations want to use this within communities without having to be kind of beholden to the the bigger players.
• 30:01
  SPEAKER_01: Um so talk to me about that.
• 30:03
  SPEAKER_01: Like, what are some of the ways that you see nonprofits, uh other impact builders like making use of this technology and applying it to their work?
• 30:11
  SPEAKER_01: Like, what are some examples of what that could look like?
• 30:13
  SPEAKER_00: Yeah, I I definitely see a lot of potentials, especially actually for nonprofit sector, because when when it comes to model users subscription, right?
• 30:22
  SPEAKER_00: If we're saying about you know large enterprises, they have the monetary power to buy this very expensive model credits and then they give it to other developers and hey, test with it, experience it, and do some product out of it.
• 30:36
  SPEAKER_00: But for nonprofit sector, obviously, like we have a limited budget and we have limited human resources.
• 30:42
  SPEAKER_00: We even have limited technical stuff that they can you know have the energy and time to figure out how do we actually deploy it, how do we actually make use of it.
• 30:51
  SPEAKER_00: So that's why we think this open source, um like open source um decision model that we recently launched is a perfect fit for nonprofit sector.
• 31:01
  SPEAKER_00: Because, you know, you don't have to always use the most complicated, most sophisticated model for everyday tasks.
• 31:09
  SPEAKER_00: For example, right?
• 31:10
  SPEAKER_00: I can think of a scenario for you know nonprofit sector.
• 31:13
  SPEAKER_00: For example, you have this kind of like donor and um donor and um, for example, um the donation site um matchmaking, for example, right?
• 31:23
  SPEAKER_00: Or you have volunteer who and staff matchmaking, or you have, you know, okay, the kind of like crisis selection criteria for what kind of people, what kind of group will get this support from the government, right?
• 31:34
  SPEAKER_00: For those, you have very clear selection criteria, and then you just need, you just need to automate it, right?
• 31:41
  SPEAKER_00: You don't need the you don't need a very sophisticated analyst to say, hey, does this work on this situation?
• 31:47
  SPEAKER_00: Does it work that, right?
• 31:48
  SPEAKER_00: It just you make just need a model to do function very fast to make the choice for you, right?
• 31:54
  SPEAKER_00: I think this is where the the the um the decent that model that we recently launched fit in really nicely for nonprofit sector.
• 32:02
  SPEAKER_01: Yeah, yeah, that's awesome.
• 32:03
  SPEAKER_01: Um, you know, I'd love to kind of like zoom out a little bit here, talk about kind of more broadly civil society NGOs and the role of AI.
• 32:11
  SPEAKER_01: A lot of the nonprofits that I work with, especially those that are kind of on the smaller to medium-sized uh nonprofits, have a lot are very curious about AI and blockchain, but also have a lot of hesitancy, anxiety about the risks, not sure if they have the capacity to manage those risks responsibly.
• 32:26
  SPEAKER_01: Um I'm curious, like from your vantage point, what do you see as the kind of real promise that this technology brings for this sector?
• 32:33
  SPEAKER_01: And what would you say to these nonprofits that are anxious about starting to use this technology?
• 32:39
  SPEAKER_00: I would first ask the question.
• 32:41
  SPEAKER_00: Like if I would, if I were to, like if you were to dive deeper in a specific um NGO case, right?
• 32:46
  SPEAKER_00: I would first ask them, what are you uh anxious about?
• 32:49
  SPEAKER_00: Like what is the fundamental disbelief or like fundamental concern that you have for using any iTools at all?
• 32:56
  SPEAKER_00: And people may I I I expect the answer could be, oh, because I hold, you know, all this sense of information, you know, I deal with refugees, I deal with political uh concerns, like people who have all these kinds of situations.
• 33:10
  SPEAKER_00: I wanted to obviously like I want to get this, you know, get these people help.
• 33:14
  SPEAKER_00: Like I wanted to find them solutions as as much as I can as fast as I can.
• 33:19
  SPEAKER_00: So I wanted to rely on AI solutions because it it seems like it can help with my efficiency at work.
• 33:24
  SPEAKER_00: But at the same time, I'm worried that what if this information gets leaked, what if uh, you know, it's get it's backfired, right?
• 33:31
  SPEAKER_00: And I think fundamentally comes to privacy, it comes to trust that whether you can trust the tool or the platform that you rely on.
• 33:40
  SPEAKER_00: So um I think that's exactly why Fluck solution can work perfectly well with this NGO sector, uh, because we don't ask them to trust Flock.
• 33:52
  SPEAKER_00: Because essentially what blockchain does is like you don't trust anybody because like it's it's it's transparent.
• 34:00
  SPEAKER_00: Whatever you have done is seen is traceable.
• 34:03
  SPEAKER_00: Like you any transaction that you made is public.
• 34:07
  SPEAKER_00: Like I don't say oh just not just because Flock says this, so you have to believe on that, right?
• 34:12
  SPEAKER_00: You believe on the technology itself, you don't believe the representative for that of that technology if that makes sense, right?
• 34:19
  SPEAKER_00: So I think um so I think if we if we start with you know red resolving the problems and concern, and then maybe we can help them to design a custom AI solution for that specific NGO.
• 34:31
  SPEAKER_01: Yeah, absolutely.
• 34:32
  SPEAKER_01: It kind of brings me back to this um idea of like trustlessness that was very present in the earlier days of Web3 and still is to some extent, um, you know, being able to trust in the kind of the code rather than the people that are operating it, um, which I think is really interesting.
• 34:45
  SPEAKER_01: And you know, we've talked a lot about the work that you're doing on the AI side of things.
• 34:49
  SPEAKER_01: You know, at the same time, I know that you're also part of the Blockchain for Good Alliance.
• 34:52
  SPEAKER_01: Um, I'm a big fan of them.
• 34:54
  SPEAKER_01: I know that they've been doing some really cool work uh with the UN as well and other organizations.
• 34:58
  SPEAKER_01: So for listeners who aren't familiar, do you mind just giving an introduction to kind of what Blockchain for Good Alliance is and kind of the initiatives they do to support their mission and drive impact?
• 35:08
  SPEAKER_00: Yes, of course.
• 35:10
  SPEAKER_00: Uh so the Blockchain for Good Alliance is also a nonprofit.
• 35:14
  SPEAKER_00: And the difference is this nonprofit is specifically backed by Bybit.
• 35:18
  SPEAKER_00: So Bybit is the number two crypto exchange in the world.
• 35:22
  SPEAKER_00: Uh well, our mission is we only have one mission, where it's like to promote blockchain technology to create tangible impact.
• 35:30
  SPEAKER_00: So people will have less misunderstanding of mixing blockchain technology and crypto industry.
• 35:35
  SPEAKER_00: I mean, every industry, as you can see, they're good size and bad size, right?
• 35:39
  SPEAKER_00: You can't just say, okay, cryptos are full of scammers, which I disagree.
• 35:43
  SPEAKER_00: The technology itself is fundamentally good.
• 35:46
  SPEAKER_00: It's fundamentally meaningful, right?
• 35:50
  SPEAKER_00: We need we need to make sure the technology is being used in a good place by good people.
• 35:55
  SPEAKER_00: That's what essentially what blockchain for good alliance does.
• 35:58
  SPEAKER_00: So we have different pill of work as well.
• 36:01
  SPEAKER_00: Um when it comes to United Nations, it goes to our government initiative where we have, you know, uh the incubation program, we have the impact fund, we have impact leader summit.
• 36:11
  SPEAKER_00: At the same time, we also have growth activities where we directly work with universities or developers or founder community, we help them to grow.
• 36:19
  SPEAKER_00: So, yeah, I would say like the blockchain for the alliance, we we try to we try uh we try to reach as many countries and regions as possible to help local communities to you know express the idea and expand the blockchain education.
• 36:35
  SPEAKER_01: Yeah, that's great.
• 36:36
  SPEAKER_01: I love that.
• 36:37
  SPEAKER_01: And you know, something that I like about both um the work that you're doing with Flock and uh the Blockchain for Good Alliance is that it's really kind of partnership driven, community-driven.
• 36:46
  SPEAKER_01: It's like multi- uh multi-sector, um, multidisciplinary, which is great.
• 36:51
  SPEAKER_01: You've worked with the UN, obviously different UN agencies, universities like Cambridge, grassroots builders, local communities.
• 36:57
  SPEAKER_01: Talk to me about kind of like why this cross-sector collaboration is so important to kind of make these technology shifts and lessons that you've learned from how to build these these strong partnerships.
• 37:09
  SPEAKER_00: Uh that's a good question.
• 37:11
  SPEAKER_00: Um So I think um it's it's hard to rely on one single company or one single partner to finish everything big project.
• 37:27
  SPEAKER_00: For example, especially, especially the impact project we're talking about, right?
• 37:31
  SPEAKER_00: Impact project, the reason why it's becoming impact project, not for not commercial projects, because less people working on it, you know, it's very difficult to get through.
• 37:41
  SPEAKER_00: You need to have people to understand local contact, you need people who actually know how to talk with local government, understand the policy, uh environment, regulatory constraints.
• 37:50
  SPEAKER_00: You also have you need to have people who understand how to design the technical product, how to make sure everything deploys implement well.
• 37:57
  SPEAKER_00: You need also people like me to, you know, advocate and communicate with different stakeholders.
• 38:03
  SPEAKER_00: You need people uh who work in the government.
• 38:06
  SPEAKER_00: So the reason why I think cross-functional partnership cost, you know, cost even jurisdictional collaboration and exchange insight is so important because you can't rely on one person to be good at everything.
• 38:21
  SPEAKER_00: But because the project itself is so important and it's so hard to finish the last amount of work, you need to rely on everybody's expertise and everybody agreed and uh being aligned on the same type, same goal, and and then you can push further.
• 38:37
  SPEAKER_00: Otherwise, because it just it takes a long time, and maybe you know you've spent more than 20 years and you only finished 50% of the the the project you wanted to do.
• 38:48
  SPEAKER_00: And if we work on it all together, maybe you can finish within two years.
• 38:53
  SPEAKER_01: Absolutely, and I think that I also you know like that it a lot of when you work with all these different partners, it makes it easier to think about how you can bring these solutions to different places as well.
• 39:02
  SPEAKER_01: And you talked about with you know the Dominican Republic, the microinsurance uh project, that you know, it's starting there, but really it could be kind of any you could grow out to other communities, right?
• 39:11
  SPEAKER_01: And it could be like this is the testing ground, we're looking to pilot it, prove that it works, and then look at how we can expand it to other community contexts.
• 39:18
  SPEAKER_01: Talk to me about that.
• 39:19
  SPEAKER_01: Like, how are you guys working to like learn from these pilots and really make sure that you know they can be applied in other communities uh if the pilot is successful?
• 39:27
  SPEAKER_00: Yeah.
• 39:28
  SPEAKER_00: So um let's still talk about the DR project, right?
• 39:31
  SPEAKER_00: Um, because that solution is um we actually received massive support from DR government.
• 39:38
  SPEAKER_00: And so that model that we have trained, it's based on it's based on real government data.
• 39:44
  SPEAKER_00: And because that data is so invaluable, and the model performance well, and and we we'll leverage the the collaborations with UNDP.
• 39:54
  SPEAKER_00: And so we are now in discussion with not only DR, but also Peru, but also Rwanda government, because they are facing the same kind of situation, right?
• 40:02
  SPEAKER_00: And then one if we prove this solution work in one specific country, and then with their with their introduction, with their uh reference, we can we can introduce that solution to a different country through the collaborations.
• 40:16
  SPEAKER_01: Yeah, yeah, that's so cool.
• 40:17
  SPEAKER_01: Um, yeah, and I'm excited to see how some of these projects um you know continue to grow and and how the pilots turn out uh as this accelerator moves along.
• 40:27
  SPEAKER_01: And so I guess like for those that are interested, that are listening, whether it's you know, builders, NGOs, change makers, civil society, or government folks that are listening in, first off, where can they go to learn more about these projects?
• 40:40
  SPEAKER_00: Yeah, so uh UNDP Accelerator Coho 3 had just opened for you know open call.
• 40:45
  SPEAKER_00: We call it open call.
• 40:47
  SPEAKER_00: Open call means that everybody who is interested in joining this initiative, they can go to the website, either, you know, propose the challenge, or you can think you can you can join as you know a solution maker saying, like, I have a solution that I've been building that I that I think this solution is good for one specific developing country, and then I want to see whether I can participate in this accelerator.
• 41:10
  SPEAKER_00: Then you can sign up and then we will do the evaluation and maybe we can work on it together.
• 41:16
  SPEAKER_01: Yeah, yeah, that sounds great.
• 41:18
  SPEAKER_01: And for like uh those that are interested in in partnering with Flock or using some of your solutions, what's the best way for them to learn more, get in touch?
• 41:26
  SPEAKER_01: And I guess to like wrap things up, is there a final call to action or note that you want to leave with uh listeners today?
• 41:32
  SPEAKER_00: I would say um well, first of all, like call to call to action.
• 41:37
  SPEAKER_00: I I would say like if there is only one thing that you you can take away from today's conversation, I think it's you need to believe in um private, to believe in that uh ownership in AI models that you could hope to contribute, right?
• 41:53
  SPEAKER_00: And decentralized AI as any you know tools that you know that it's it's it's it's like it seems impossible, like it seems like so new, like you know, seems like so not like not mainstream, but it will become mainstream, you know, eventually will it will.
• 42:11
  SPEAKER_00: It's like internet, right?
• 42:12
  SPEAKER_00: In the beginning, the internet is closed, and then now it's like open to all, right?
• 42:16
  SPEAKER_00: And we think the AI should be the same thing, like the AI model should be owned by everybody as well, not just by big tech.
• 42:22
  SPEAKER_00: And um, and that's like I think like one of the calls to action.
• 42:27
  SPEAKER_00: And how do we get in contact?
• 42:29
  SPEAKER_00: I mean, you can find us in all the social media, you know, we're in LinkedIn, we're in X, we're in Instagram, we're in YouTube, just um, or like we also have our website as well.
• 42:39
  SPEAKER_00: You can you can write us an email.
• 42:41
  SPEAKER_00: And for you to join Accelerator, you can just go and search for UNDP Accelerator Cohort 3 and then to join the open call.
• 42:49
  SPEAKER_01: Amazing.
• 42:50
  SPEAKER_01: So lots of ways to get involved, at the very least, lots of ways to learn about the great work that's being done here, to gain some inspiration if you're working in you know the NGO or impact sector and want to see how you can learn from some of these solutions.
• 43:01
  SPEAKER_01: So I'll make sure to include all those links in the show notes, of course, for those listening along.
• 43:05
  SPEAKER_01: Um, Tiffany, it has been an absolute pleasure.
• 43:07
  SPEAKER_01: Thank you so much for taking the time to be here with us today and for sharing your wisdom.
• 43:10
  SPEAKER_00: Thank you so much for having me.
• 43:12
  SPEAKER_00: It was a pleasure.
• 43:14
  SPEAKER_01: And that was Tiffany Wang of Flock.io and Blockchain for Good Alliance.
• 43:18
  SPEAKER_01: What an awesome conversation.
• 43:20
  SPEAKER_01: Uh, great advice for nonprofits in there.
• 43:22
  SPEAKER_01: Uh, some incredible case studies, and I just love seeing innovative projects like those we discussed in this episode coming to life all over the world.
• 43:29
  SPEAKER_01: There are a couple things I want to highlight that stood out to me, and the first is the importance of data sovereignty.
• 43:34
  SPEAKER_01: For the communities so many NGOs serve, data can be everything.
• 43:37
  SPEAKER_01: I mean, we're talking health records, financial histories, locations, identities, and in the wrong hands it can cause a lot of harm.
• 43:43
  SPEAKER_01: The idea that communities shouldn't have to give up control of their data to benefit from AI is a principle every organization in our space should keep top of mind.
• 43:51
  SPEAKER_01: The second is how practical these pilots are in real everyday needs.
• 43:54
  SPEAKER_01: Whether it's the projects getting aid to people in near real time or those with options for feature phones and even paper vouchers, or helping farmers, artisans, and conservation communities capture more of the value they create, they aren't necessarily flashy experiments.
• 44:07
  SPEAKER_01: They're practical solutions to problems that have frustrated the development sector for decades, and the partnership with UNDP shows how seriously the international community is starting to take them.
• 44:16
  SPEAKER_01: And the third is the power of collaboration.
• 44:18
  SPEAKER_01: No single organization can make the shift alone.
• 44:20
  SPEAKER_01: It takes the UN, universities, governments, NGOs, and grassroots builders all working together, which the NDP accelerator really exemplifies perfectly.
• 44:28
  SPEAKER_01: For the NGOs, nonprofits, and civil society folk that are listening and are curious about AI and blockchain but unsure where to begin, I hope this conversation gave you both a sense of what's possible and some confidence that it can be done responsibly.
• 44:39
  SPEAKER_01: And if you haven't yet, check out our last episode where we highlight some more uh case studies and nonprofits using AI successfully.
• 44:45
  SPEAKER_01: And you know, the important takeaways I want to share here are one, start with a real problem.
• 44:50
  SPEAKER_01: It's important that you really have a problem that there is a use case that there is a need for, as opposed to just seeking out AI for the sake of it.
• 44:58
  SPEAKER_01: Uh really keep your communities at the center, and look for partners who share your values along the way.
• 45:02
  SPEAKER_01: So, for listeners who want to learn more about Flock.io, the UNDP SDG Blockchain Accelerator, or the Blockchain for Good Alliance, we've included all the relevant links in the show notes.
• 45:10
  SPEAKER_01: A huge thank you to Tiffany for sharing this work with us today, and to everyone at Flock.io and the Blockchain for Good Alliance for helping build technology that puts community first.
• 45:20
  SPEAKER_01: And that's up today's episode of CryptoAltrist.
• 45:23
  SPEAKER_01: Thank you so much for tuning in today.
• 45:24
  SPEAKER_01: I hope you enjoyed the conversation as much as I did.
• 45:26
  SPEAKER_01: If you're inspired to explore more stories at the intersection of Web3 and Social Impact, be sure to visit us at CryptoAltris.com for articles, resources, and more.
• 45:34
  SPEAKER_01: And if you love what you heard, it would mean the world if you could rate, review, and subscribe to the podcast.
• 45:38
  SPEAKER_01: Your support is huge in helping us reach more listeners like you and keeping the platform going strong.
• 45:42
  SPEAKER_01: Thanks again for joining us, and I look forward to having you back for the next episode.
• 45:45
  SPEAKER_01: Until then, let's keep showing the world the good of crypto.