EP 122: Advancing Equality through AI and Cutting-Edge Technologies with Zameer Brey

This is the second episode produced under the #STEMSolutionsAfrica series. The guest today is Zameer Brey,  the Interim Deputy Director for Technology Diffusion at the Gates Foundation. He gives us a rare glimpse into the backstage of technological revolutions and how AI is breaking barriers to access vital resources in developing nations and marginalized communities.

Twitter: @ZameerBrey
LinkedIn: Zameer Brey
Facebook: Zameer Brey

Transcript

Anne Chisa: 0:15

The Root of the Science Podcast with your girl, Anne with an E. Hello everyone and welcome back to another episode of the Root of the Science Podcast with your girl, Anne with an E. If you are new here, welcome, welcome. It's always such a pleasure to have new listeners to the show and I hope that you become a regular. And to our regulars, man. As always, it's such a pleasure to have you returning. Thank you so much for coming back and listening to the show. Now a reminder that you can follow this show on the various social media platforms. You can follow on Twitter, instagram, tiktok, at the root of the side pod, or you can follow on Facebook and LinkedIn at the root of the science podcast. Or you can also listen to this podcast wherever you listen to your podcast, so it can be YouTube, it can be Apple podcasts, it can be Spotify or even Google podcasts, and make sure you hit that notification button or subscribe button specifically on YouTube. Now let's get into today's episode.

Anne Chisa: 1:22

In Africa, artificial intelligence and other technologies are playing a crucial role in improving healthcare, food security, agriculture and education. These innovations promote equitable access to essential resources and services across the continent. Today we are speaking with Zamiya Bray, the interim director for technology diffusion at the Gates Foundation. In this episode, zamiya will tell us about the foundation's efforts in implementing AI and other innovative technologies across various sectors. He also shares some real life stories on how the foundation's efforts in developing technologies to improve healthcare, for example, has enhanced access to people in developing countries and marginalized communities. Additionally, he shares with us some exclusive details about recent projects initiated by the Gates Foundation. Tune in for this and, of course, so much more. Let's go. Good morning, zamir. Welcome to the show.

Zammer Brey: 2:23

Good morning and thanks for having me.

Anne Chisa: 2:26

It is such a pleasure. I'm so excited to chat to you this morning and for the viewers to hear all about the amazing work that you do. So to just get things started, could you just tell us more about your role as the interim director for technology diffusion at the Gates Foundation?

Zammer Brey: 2:50

Sure, and let me just thank you for leading this wonderful initiative that allows us to put the spotlight on science and the exciting world of innovation. So at the Gates Foundation, we've increasingly recognized the importance of understanding the problem and the context we're operating in, so that we have a much better fit between the solutions we're developing and the challenges that communities face on the ground. And so my role, in the simplest form and is just trying to join dots, nothing more complex than that, but trying to join dots in meaningful ways that allow us to develop solutions that better fit the problems and challenges and context we're operating in, and so making sure that we have a better fit between these solutions and there's a better pathway to scale up these solutions. And so a lot more intentionality around understanding context, communities kind of stake, all this perspective so that the solutions that we ultimately invest in have a higher likelihood of successful uptake. So that's the motivation behind the role, and it is intended to work across multiple technologies, but for obvious reasons, it'll be focused on AI for the near future.

Anne Chisa: 4:34

Because that's where the world is going to right now.

Zammer Brey: 4:37

Right.

Anne Chisa: 4:38

Yeah, you mentioned that you joined dots to develop these solutions, so could you take us to the beginning? How does the Gates Foundation identify a problem, right, and then how do they select, for example, some technologies to invest across the various sectors that you work in?

Zammer Brey: 4:57

Right, that's a great question, anne. So just to underscore, the focus of the Gates Foundation is to work to reduce inequities. We believe that everyone should have the chance to live a good, healthy, productive life, and so we are working in areas and communities where that isn't happening because of some other phenomenon. Now, that could be health related, it could be malaria, it could be tuberculosis, or it could be related to economic well-being, agriculture or the marginalization of women, and so we really work to reduce inequity in the world. So our starting point is really on the why. Why do we do what we do? Then we try and understand to what extent is the market failure in a sense, in that particular community or geography? So, just to break that down, there are many areas of the world or communities that are affected by a particular disease or phenomena, but actually the market generates those solutions because there's some commercial value. We would operate where the market itself is not going to solve that problem.

Zammer Brey: 6:29

And a good example, anne, to talk about is tuberculosis In our country today, an aeroplane full of South Africans will die every day because of tuberculosis in South Africa alone. So I'm just going to pause and repeat that and it's an aeroplane full of South Africans, of people living in South Africa, that will die every single day. Now just imagine that happened today an aeroplane crash. If it happened tomorrow, I'm pretty sure the president will call for an inquiry.

Zammer Brey: 7:10

If it happened on the third day, we likely will see the airspace close down right, but tuberculosis is a disease that does proportionately affects the poor, but 100 years later, there's still is no better vaccine than good old BCG, which has very limited efficacy, only protects people in a certain age group and doesn't work as well.

Zammer Brey: 7:41

And so it's taken us 100 years to get to the point where now, at least, we have some exciting news to share, which is we have a very promising candidate that will go into a phase three trial and that was just announced last week a 500 million grand trial sorry dollar trial sponsored by the Gates Foundation. Welcome to us and several other partners, of which South Africa is going to play a leading role in driving this trial, and we hope that this brings ultimately a vaccine to prevent the onset of tuberculosis infection and onward transmission, which will primarily serve the poor and reduce inequity. And so that's a really good example of how the foundation's coming into a space where the market otherwise wouldn't produce a vaccine. It's going to be really expensive, we have to take a long game, but if we believe that the foundation played a catalytic role in getting to this point, having invested in more than a dozen vaccines over the last 20 years, that's absolutely, absolutely fantastic, and when you use that analogy of an airplane, I think it puts things into a lot of context.

Anne Chisa: 9:05

I think sometimes with some of these things, like you said, that B affects marginalized communities. I have to say I don't even think twice about it, but there are so many people who are affected and it's really great that, as a foundation, there is a charge being taken to solve this very prominent problem, and I really am excited to hear what happens after the first trial in terms of the next steps. That's really, really exciting and I commend that work.

Zammer Brey: 9:39

Thanks, anne. You know tuberculosis is one of those phenomena where and there are other disease and phenomena, but it really has the potential to be catastrophic, not just for the individual but the family. I remember this really touching story and, if I may, for a minute of meeting Cipucazi in Kailitsha and she told us her story of how she started coughing, losing weight, was really unwell, went to the clinic not once, twice, but thrice, and then got diagnosed with TB, was started on medication. Now at this point just to say she had a stable job, was working at Canal Walk, a very big mall, had a daughter that she was caring for but was starting to feel unwell and was taking some sick leave to attend the clinic.

Zammer Brey: 10:49

The medication didn't work and unfortunately it turned out that she had drug-resistant TB and five years ago drug-resistant TB was extremely hard to cure and in many situations had a worse outcome than most cancers in the world. So less than 40% of people would survive drug-resistant TB. That feel has rapidly advanced. Cipucazi was one of the first people that got the new drug-resistant TB regimen in South Africa, thanks to the Department of Health, msf and a bunch of partners that really made that accessible. But here's what happened to her. Her employer led with stigma and said even after she was considered safe in terms of coming back into the work environment, her employer fired her just on the basis that she had a diagnosis of tuberculosis. She no longer had an income. Her daughter suffered, she went through most of a drug-resistant regimen alone, isolated, without a regular source of income, and so it's really tragic to see how a single diagnosis can destroy the livelihood of an entire family in a matter of days and weeks.

Anne Chisa: 12:18

That is so sad and I just want to pull the string further on that, and to touch on accessibility.

Anne Chisa: 12:24

You spoke about that for people who are like Cipucazi, for example, who are in marginalized areas, and I want you to just explain further that obviously there is a bit of a digital divide. For example, in some of these marginalized areas, they don't even know that these things are actually available or that, for example, the Gates Foundation is now doing a third trial. So how do we make these technologies accessible and also, most importantly, affordable for somebody, so that these stories don't become so common? Because, if we're being honest, they are right.

Zammer Brey: 13:04

Absolutely, absolutely so, I think, and I would say, and we haven't solved this yet. For us, this is a journey, and we want to make sure that we're solving it in a way that enables access to the most vulnerable populations, and we do that in a few different ways, and I think we'll touch on that specifically around the AI work that I'm leading on when we get there. But we work with manufacturers and scientists that discover these technologies to make sure we have global access, and what that means is we want to make sure that whatever technology and innovation we, as the Gates Foundation support ultimately can get into the hands of the people who need them and will benefit from them the most. We know the world has a history of releasing incredible technologies that will go to those who have the most money in their back pocket. Sadly, we saw this play out in not too distant past with COVID, not only globally, and there was some really smart work done in South Africa that demonstrated that, even within a country, the amount of money you had in your back pocket determined how quickly you were going to get a COVID vaccine, and so you had persons who had medical insurance getting a vaccine at five times the rate of those who did not have medical aid, and then, obviously, if you break that up and you kind of have what looks like a social gradient of access to life-saving treatment.

Zammer Brey: 15:01

And so what we do is we work with manufacturers, number one to make sure that those products become available as soon as they are discovered, are safe or effective to the countries that will benefit from them the most.

Zammer Brey: 15:17

The second thing we do is we try and create pathways, working very closely with the governments and regulators, to make sure that, as these products become available, there are the mechanisms, supply chains and ways to get them to the most rural, the most distant communities.

Zammer Brey: 15:39

And then we work with social enterprises and NGOs to enable individuals to get those medications in a way that's accessible, easy, convenient. And sometimes the assumptions we make as folks working in development around oh we've just got to make this available and everybody will come are fatally flawed. And so we, increasingly, are moving how we think about access to something that is so convenient and simple and easy and that strips out the stigma that often exists in our health system or in some of the other areas. We work to put the user at the center of what we design, and so we will often do market research with end users who are affected by these diseases to tell us how do they want to access this, what to make the most sense for them? Often that means we've got to break our mental model of health care is only when you come to a clinic or hospital.

Zammer Brey: 16:51

How do we deliver your health care in a way that is simple. Could it be at a taxi rank, could it be at a bus station, could it be at the post office, could it be at the retail center? And so those models of care are what some of our teams are starting to think about, and again, we haven't arrived, and it's only because our partners on the ground that we fund and support are able to shift the boundaries and make sure that we're delivering things in a way that's user-centric, rather than relying on people to come, which often. Again, we have 20 plus millions of Africans in this country, in South Africa, that rely on a social grant For them to take transport. To come half way across the city to collect a medication or get a vaccine is actually an unfair cost. It's relying on somebody to use 20 or 25 percent of their monthly income to come and get that. And so we've got to think about new models of K as much as we think about new products and new services.

Anne Chisa: 18:14

Fantastic. I like that. New models of K is really, really important. So now let's talk about these technologies that we are trying to tell people about, and I'll be quite interested. What are some of the specific AI-driven technologies that the Gates Foundation has invested across sectors? You mentioned that you work through health. You work through agriculture as well, for example. So how is AI being applied? Because I'm really excited to hear about this, because that's the new buzzword now.

Zammer Brey: 18:50

Absolutely. And again, thanks for the opportunity to talk to you and because I'm confident that this discussion that you and I are having here is going to get to some of the innovators that have actually come up with these amazing ideas. And so, if it's okay with you, and I'd like to talk about the Grand Challenges Program, which is our first step, and, we hope, in the right direction. But there are some real signals that tell us this has met the need of young, creative innovators, and so, just to preface that, the Foundation has been involved in some AI solutions before. Generative AI's bubble has expanded this year, but a lot of that has been involved on the imaging side, some voice to text work, but this year, an explosion of Generative AI tools.

Zammer Brey: 19:54

That gives us hope that we can leverage these for development, and so the sharp edge of what we're doing here is we see this as a force to reduce inequity, and one of the ways we want to do that is by building out the evidence base, and so, just as we started off the discussion is how do we know we're doing the right thing? Well, let's lead by asking communities to tell us what they think the biggest problems and priorities are where they are. So we did not specify a problem statement. We basically invited applications across all the development sectors we work in, which include global education, agriculture, sanitation, health, financial inclusion, and we said to communities, partners, ngos, for private companies tell us which problem you are working with in your geography and submit a proposal. And we were absolutely blown away. Yeah.

Zammer Brey: 21:05

We had. We gave people a very short time to apply Only two weeks. We had 100 applications every single day.

Zammer Brey: 21:16

Wow On average, we ended up with 1300 applications in that period of time and years, years. The exciting piece 50, more than half of those applications came from this beautiful continent, africa. So more than 500 applications came from Africa. But the broader pool was from 103 countries across the globe. And as we looked through the applications, they were triaged, then they were reviewed by a set of expert reviewers. In kind of 10 days there was some 600 hours spent triaging and re-reviewing these applications by 80 reviewers and another 10 triage folks. The bottom line is the quantity of applications excites us but the quality of the applications inspires us.

Zammer Brey: 22:26

We were just blown away by the thinking, the foresight, the vision of these applications, so much so that we are internally discussing increasing the overall envelope of funds that we've put aside. So we committed to funding 30 applications, but we saw many, many more fundable applications, and so we will not make that announcement imminently. But I just want to give you a heads up.

Anne Chisa: 22:55

Exciting.

Zammer Brey: 22:57

To double our envelope.

Zammer Brey: 23:01

In addition, we wanted to hear from new partners. We wanted to diversify the grantee base, because we believe that AI doesn't need people to be experts in AI and I'll tell you one or two stories and so we put this application as far and wide as possible, and the result is that 75% of our grantees are new applicants to the Gates Foundation, new recipients of funding to the Gates Foundation, and that, too, excites us that we're going to be hearing from new voices, new partners, new communities, and we think that it's an expression of the creativity and energy around leveraging generative AI for development in ways that we are ready to support. And this is just the first step of many to come.

Anne Chisa: 24:09

That's so, so exciting. I also like the fact that you said you made it very broad and also quite exciting that these are people who are not so much as experts, and that way I think sometimes there's a bit of gatekeeping when we only look for inverted commas experts in a specific field, because they're people who are very smart and they learn in various different ways. So it's quite exciting to hear that you've got such a wide range of people that have applied. So once this application process goes through and the recipients are informed, what happens next?

Zammer Brey: 24:58

Yeah, absolutely, and so I'll talk to that in a second. And I just want to say that this is also an opportunity for the Gates Foundation to create a platform to amplify the voice and give agency to partners in the global south. We recognize that, if this technology is just left, that organizations in the global north and commercial entities will use this tool, and we are seeing that. But our role is to try and use this as a tool to reduce the inequities and, by creating a platform for partners from the global south, to really leverage this, and so over the next three months, they will be taking their idea that they've defined in their community and running a proof of concept. Now what's becoming increasingly important in this space is that we've got to find a way to cut through the hype and channel that energy to strong evidence generation. We can't be in a space where we're taking very big decisions based on anecdotal kind of things that look cool and feel cool but don't have the same level of evidence that we would expect for making big policy Decisions, and so what we want to contribute to in one way we have some other initiatives, but in one way is to build out that evidence base in a way that's underscored by a responsible, ethical, transparent, traceable, you know evidence generation. And so we're doing something special with this cohort and, to give you a heads up, we actually haven't even shared this with grantees, but I'm happy to share.

Zammer Brey: 26:58

This year, because we will announce it next week, we're going to provide this cohort with extra support. So we have a network of researchers from Africa, south America and India that will provide specific expertise on methodological strength for AI, on gender inclusiveness for AI and on ethical approaches for AI, and so this cohort of grantees will have access to untapped kind of untapped expertise from, you know, from the field, so that they can strengthen the approaches and proposals. Then, after the three months, we have, you know, now, some interesting discussions about so what you know. Three months is really not enough time to really robustly test something. Our sense is that from this cohort, a number of these proposals will show promise, the proof of concept will be positive, and so we are just, you know, exploring ways now to continue that funding for longer term, and that's to be determined and announced, but we know that there's going to be some interesting results here. The second piece is that a number of our fellow donors and in-country partners are considering making kind of more funding available along this stream of proof of concept testing, and so I should have said each of the recipients will receive up to $100,000 for the three month period, which we hope is enough funding. We think it will be enough funding to provide support for that initial period and then, you know, some more kind of support beyond that Notwithstanding, and that there are some, you know, fairly significant structural constraints that we're working against in Africa, and so we certainly don't want to ignore that or, you know, be like the Austerians deco-eating the sand on this one, because those are real and those are challenges that we want to work with partners to figure out how to unblock, and we have to acknowledge that we're operating in a space with a real structural constraints.

Zammer Brey: 29:35

Right, you and I are both experiencing low-chaining today. Electricity supply on the continent is real, you know we talk about infrastructure and you know the funny thing is I was being interviewed for a large media publication and my lights went out, so I was standing in the dark and we were having a discussion about how exciting the AI developments are going to be, and so it's real.

Zammer Brey: 30:05

And I think electricity supply is one issue. Bandwidth is another issue, not only the cost of access to internet. But even in our country, I was just shocked to learn that only 7% of children in South Africa have access to reliable internet. Now, it's one thing saying AI is going to solve our global education crisis, For who.

Zammer Brey: 30:42

Is it going to solve it for upper and middle income folks or is it going to solve it for those that need it most, you know? And so that's another challenge, and I think we need to be honest that there's a skills gradient that we're dealing with. So, to give you one example of this, africa has the same number of software engineers as a single state in the United States of America. So the California state has the same number of software engineers that are critical for AI, foundational building, api build, integration, all of those things as the entire continent, and so there's a real kind of inequity of skills. And so, as we think about what's this next stage of work, we've got to think about the ecosystem elements. That certainly not the foundation's role to do that alone, but certainly something we've got to go in eyes wide open and acknowledge and be cognizant of.

Anne Chisa: 31:55

And it's important to have these conversations in the beginning, to know what we're getting ourselves into, because sometimes we don't want to be caught with our pants down down the line when we we already know these, these types of issues, and I think it's so important to speak about it. Electricity, internet, and these are things that, for example, somebody in the middle to upper class takes for granted, but the people who, as a scientist, I know we create, we do science to solve societal issues, but we can't solve societal issues when all of these issues are structural are there and it's important that, as we navigate through our development of AI and all of this new way of thinking, that we keep that in the back of our mind, and I hope that, as the grantees who are working to solve these various solutions, they also think about that when they're implementing these different types of solutions.

Zammer Brey: 33:01

Absolutely, for many of the technologies we invest in, we expect career scientists, researchers that have many, many letters behind the names and degrees to be leaning in and solving these complex things. But we are seeing with AI the genitive AI, that is an opportunity for the Africa youth population to lean in. Now we know that Africa's got, you know, a high proportion of energized youth that want to make a difference, and so it doesn't necessarily require somebody that's got deep expertise to try and use some of these tools. Recently my son asked you know to for my laptop. He needed to give a farewell speech at his favorite aunt's farewell. She was departing and he asked for my laptop and I thought he was just going to use word and you know, type up, his thing is 11.

Zammer Brey: 34:12

The next thing I saw him he was using chat GPT and he was actually prompt engineering. So he gave chat GPT an input and he said here's my favorite aunt. This is what I love to do with her, can you type something up? And it came up with a long speech and I kind of just stood back and watched this. The next thing he said can you shorten it? And, by the way, I don't understand these three words. And can you add a joke to? And I watched this and there was a mini prompt engineer in him, you know. He then read it, wow, to an audience that will absolutely bowled over. And my dad was an orator kind of looked at me and said, gosh, your son reads so much. And I said, no, he doesn't read so much, he just is leveraging chat GPT Right.

Zammer Brey: 35:08

Now we can see that as a threat in our very traditional education system. Or we can see this as an opportunity.

Zammer Brey: 35:16

It's going to be a. It's pervasive. How do we get kids to use this creatively, safely, responsibly? But you know, I'm sure if I let him continue to play with it I mean he's now written a poem about his chickens in the backyard. He's got all their names and he put it in and gave him a beautiful poem. That's how he knows them. You know all their names and so I just think there's a moment here, there's a step opportunity for us to give you the space to solve problems that they are being affected by, but also that our communities are being affected by, and so we're putting it out there that it's a space where we are increasingly thinking about is how to leverage this kind of you know, youth population on the continent to help us.

Anne Chisa: 36:12

Wow, this is such an amazing story and you know, I think it's so important that you've brought that up because I know sometimes with AI and chat, gpt which is the most common one that everybody knows there's that whole idea that it might be a threat to the education system. But I think on a personal front.

Anne Chisa: 36:30

These are my personal opinions. I think it's about how you use something and it's not to minimize education or to make because there's that chat like, oh no, it's going to make kids dumber, but I think in a way, it makes them very smart and it makes them think in a very in a different way and a non linear way and it makes them think out of the box. So I think it's how we leverage AI and not to say that we dismiss the education system, but it's really exciting to see your son as he live it, imagine how he's going to be by the time he's 16. And what he can do, because it also makes it so exciting for the kids and it makes it like, wow, I can create a whole poem or I can create a speech about my favorite orange.

Anne Chisa: 37:16

And that makes them excited to actually do this some more right.

Zammer Brey: 37:20

Absolutely, and so I think there's just a myriad of opportunities that have opened up that we have to embrace. You know, obviously we've got to do it in a way that's responsible. So, for example, we have to acknowledge that there are many languages on the African continent that are not properly set up and structured for right now. We also have to acknowledge that there are serious biases in how the current set of AI tools portray women, marginalized communities that don't have a voice, and I think we've got to approach that with a level of caution. At the same time, we should be asking ourselves so what? How do we?

Zammer Brey: 38:08

you know, leverage the tool today in a way and on a set of use cases that make the most sense for our population and continue to try and weed out the biases that are problematic.

Anne Chisa: 38:24

Exciting. This has been such an exciting conversation, sameer, and just as your role as the interim director of technology diffusion. I think when you touch on your personal stories, it really goes back to what you said about how your job is to connect the dots and it starts from the beginning, with your 11 year old son to what you're doing in your work life, and it's been a pleasure chatting with you and I hope somebody is inspired and is also excited about the potential that AI has, especially for us in the global south, especially for us and in the African continent. I'm really, really excited about what's coming, what's really coming. So thank you so much for sharing with me today.

Zammer Brey: 39:16

Sure, and you know again us and Sia. Thanks to you for creating the space for us to share a little bit about the path we are taking and the stories that are emerging. And you know, just to say, after having looked at many, many applications, that have come through from, you know, partners in India, Bangladesh, Pakistan, Ghana, Senegal, Rwanda, South Africa, et cetera.

Zammer Brey: 39:46

I kind of left with a feeling that right now, if we thought of AI as some canvas, the only border to that canvas is actually our creativity, and so we've got to figure out how we use our creativity for good, so that, you know, ai can become a force equalizer and reduce some of the inequity we see across the world, and we are optimistic about that journey and thankful to partners like you that you know, share this vision and create the space for us to share our story.

Anne Chisa: 40:28

It is such a pleasure, and to everybody else who's tuned into this episode, thank you so much for listening to another episode of the Route of the Sons podcast with your girl and with an E. Until next time, goodbye.