Summary

What skills will our students genuinely need to thrive in a future driven by AI? To find the answer, Daniel Emmerson goes straight to the source and sits down with brilliant young minds behind seven teams from the Hult Prize Global Accelerator, one of the final stages of the world’s largest student startup competition.

This episode takes you on a global tour of innovation. You'll hear how these young entrepreneurs are using AI to tackle large problems, like enhancing public safety by turning CCTV cameras into proactive witnesses, helping firefighters respond faster, pioneering sustainability by transforming agricultural waste into valuable resources, and creating a "calorie counter" for your carbon footprint. The conversations also cover how AI is being used to deliver personalised education in Ethiopia and provide gentle, effective speech therapy for children.

Although these startups focus on very different issues, they all agree that the future isn’t about AI replacing people, but being empowered by it. Tune in to find out the essential skills our future generation will need.

Transcript

Welcome to Foundational Impact, a podcast series that focuses on education and artificial intelligence from a non profit perspective. My name is Daniel Emmerson and I'm the Executive Director of Good Future Foundation, a nonprofit whose mission is to equip educators to confidently prepare all students, regardless of their background, to benefit from and succeed in an AI infused world.
‍

Think Hub ET Innovations

Daniel

Welcome to Foundational Impact! Gentlemen, it's wonderful to have you here. Um, our podcast series, as you know, explores artificial intelligence in schools, predominantly where we work with teachers in trying to understand what responsible use of AI looks like in schools today, and of course for the next generation. So through this collaboration with the Hult Prize, we're trying to understand how startups are thinking that they might use AI or tech strategically within their organisations so that we might better prepare schools to backwards design the skills that young people will need as they go into the world of work in the future.

Think Hub ET Innovations

Thank you for having us, Daniel! Our company is Think Hub Innovation. So it was founded, uh, three years ago on the purpose of, uh, delivering quality education to Ethiopian students and mostly East African students. We came up with this solution because we've been the victims of, uh, the quality education gap. So, uh, I have been learning, uh, working long distances too. Schools getting guidebooks only for one hour, and I have to pass it along to the next student.So it was just a race against time and I did not have that much of resources. And as of my co-founder, I would say we same, the same challenge. So we want to change that for, uh, Ethiopia and mostly South African countries that are suffering from this.

Daniel

How, what does that change look like for you? How are you addressing the change that needs to happen in, in Ethiopian schools?

Think Hub ET Innovations

So like currently in Ethiopian education, the teach, the teaching method is more outdated in the resource materials, very in the school. So anybody communicate any or educational material easily. So with platform like us, it's easily to get these kind of materials on their faults. So in addition, I would say, uh, smartphone usage coming very high, but the teaching method still remains the same.

Now we are trying to fill the gap between students having access to smartphones. And the problem that we give, uh, the teaching method that we give is so outdated. So we wanna match that and give the students that the proper, uh, access team meet. 

Daniel

Your main product is a platform, is that right? For uh, for accessing educational content?

Think Hub ET Innovations

Exactly, yes. Via mobile phones. Yeah

Daniel

Can I ask a little bit more about that Please? The, the AI that you're using, the AI that perhaps you are, you are building or working with. Can you tell us about what that looks like and what their experience might be from a student perspective?If they're engaging with it? 

Think Hub ET Innovations

Great. Uh. We are using, uh, open source, of course, API as to be specific GT four. And on the student perspective, sorry. Well, one is the creating the, uh, quizzes. After creating that adaptive quizzes, then it'll go through, uh, checkup by the instructor, the teachers, and we would be delivering to the student.

And the next thing is personalization. So if a student, uh, needs to improve one subject and, uh, the student is good in another subject, like it would try to do the notebooks and, uh, be personalized as, uh, possible. Maybe add that. Yeah, like as you said, it's more. Like we trying to connect to the s like to see where they are right now or when they step up, when they start scaling, we'll take some kind of test runs where the students will at and after that the AI will give them where they are anyway, they own put, and which subjects they want to stab or to focus on, which these are very low and embeds, and which they are on high grades. So my analysis is streamless. The AI will give them. Which part of the subjects they want to work on. Focus on.

Daniel

Final question from me then, gentlemen. What advice might you give a student at school today who is thinking about whether or not they need to be learning about responsible use of ai, even while they're still in, in middle school or high school?

Think Hub ET Innovations

Uh, they really need to be careful on the use of AI. Uh, I will make, uh, doubles, uh, I would make it very, uh, a remark when I taught this because, uh, if you use it for bad things, like to do your assignment or something like that, if to, it'll really kill you. It's just, uh, you have to use it properly so that you become effective at not doing your job.

AI must be helping you and try to look at, at broad art. It could be it. Don't let it replace you. That's my final statement. Yeah, like it's the same here. Don't let it be replaced. Focus on what you are doing. Like focus on what you are studying. Try to think outside of AI and ask AI to help you get there what you want, what you wanna achieve. So don't let it replace you too, yeah. 

Daniel
Thank you so very much. Excellent. Thank you.

Interlinked

Internlinked

What we're aiming to do here at Interlinked with our algorithm is. Uh, right now in the fire response, uh, situation in California, it takes 20 minutes for departments to go in and fight a fire. So we built an AI algorithm to not only stop that 15 to 20 minute latency gap, but give them response strategies utilizing AI in real time to make their decisions better. 

Daniel

When you're talking to firefighters about this and the work that you're doing, I'm interested to know as to whether or not the aspect of trust has come up at all in terms of trusting the data and trusting the, the decisions and the requirement for input.

Because there's a lot of conversations, certainly in schools around AI being able to provide advice or recommendations or guidance, and then acting on that guidance is like the next step, right? Taking it into the real world. Can you talk to us a bit about that?

Internlinked

The biggest thing is when we first went to them with AI, they were very. Pushback. It's a new, daunting concept, especially for an industry that's on the ground. How can technology potentially help us?

The biggest thing with them was teaching them that this is not here to replace any of their thinking. It's an enhancement. Right? So kind of explaining it in a simple way to say, Hey, whatever you guys do, your decision making is up here. We're just giving you everything on the table in real time to make that decision a little bit easier.

So that eased that stigma a little bit, and once we got to that point, it allowed them to really start trusting it. Um, that's, that's been the case so far across California. We started testing in a couple counties outside, but within California there's been a huge media pushback finding everyone's looking for new technology to help the fires, especially what happened.

In the Palisades in January. Yeah. Brutal fire. Um, down in Southern California. So there's been a push to look for a solution. So there's a, it was a little bit easier than we anticipated, but as I said, to make this an industry stalwart, it's all about trust. Making sure that they're comfortable using the technology to talk about pitfalls we've had when we first started generating responses through these pilots.

The outputs we were getting were very long dent. Every detail of the AI would identify every single thing, right? So we realized they can't digest that in the moment. They need something simple. So we've now designed it to be a simple convertible AI teammate they can chat with, ask the questions that they want to ask.

So really the most important thing is working with them and learning what they like, what they don't like. That's really the purpose of our pilots. Right? And that really came over time because when we first started this. We just wanted to know what types of problems they actually face. And a lot of firefighters actually use Google Maps or even physical mapping to actually draw what their responses were. And you know, we were taken aback because. The, the lack of technology in the space is just a little mind boggling to us as we're all technical, you know, so and so on.

Daniel

Stepping away from the work that you're doing at the moment and thinking about the next generation, the gen generation who are still in, in K-12 education at the moment, would you have any advice or thoughts, particularly for teachers who are thinking about how they might look at skills young people might need when they're coming out of school and, and looking at university and the workplace?

Internlinked

Yeah. Uh, I love this question. So at Berkeley, there's a program which, uh, allows current Berkeley students to teach high schoolers and teach. People, uh, that are younger, usually age between 13 to 17. So I actually did teach one of those classes. And, um, coming, having both sides of it, being a student, but also teaching, I've noticed that AI coming in is something new.

You have to learn to balance it, right? So allowing them to utilize their skills to whatever they think AI can help them with. So be it. But as long as it's in the boundaries of, you know, uh, it's to enhance your learning, you're not using it to replace concepts. You need to keep making those same. Uh, important claims, right?

You can't just use it to like, write this essay and just forget about it. You know? So I think that's something that I've realized. And having those conversations with kids, they'd ask me questions like, uh, do you think I can do this to do my entire task? Or how do you think I should utilize it? I tell them the same thing.

I feel these guys, when we're working on the startup, whatever you think can enhance your understanding or allow you to get more ideas, use it to brainstorm, use it to throw ideas back like it's a, a person, but it has access to the entire internet, the entire scape. Using it like that, I think was what the main point came across as a research tool really.

So do you have anything? Yeah. Um, AI applied to EdTech is a really big upcoming space and I definitely think that honestly it should be a tool like Han mentioned that is used to boost productivity and learning, not necessarily replace it. Right. Um, and I think that's a big stigma nowadays that, you know, like chat, GPT can just write essays.

It can do your math, right? Probably wrong, but it'll do it right. Um, but yeah, I think it's more of a tool that you use to basically explain things deeper. Uh, explain concepts, get a better understanding and a deeper understanding of different things that you're learning. Yeah, and I, I think whether we like it or not, AI is definitely going to be a part of our lives in ed tech and any other industry.

And they've already started to encroach into Berkeley college classes. Yeah. Some professors have told us. Hey, just use AI, uh, for some tasks if you feel it's menial. So, you know, really getting to know how to use it responsibly is the, the biggest important factor. Yeah, and that's why, just my last point I would say is I've had professors tell us.

Just completely use it, completely stop using it. And that leads to a stigma with a lot of students I've seen where they're like, okay, now we gotta use it. I don't know what we use it for. I think the better approach to look at it is when you have students utilizing ai, 'cause it's coming, as Neil said, whether we like it or not, to teach them the best ways to use it in that scenario.

So for example, in the classes we taught, it was all about use. Utilizing as a research tool, find the market size, allow it to give you data that you have to then analyse yourself. It allows that productivity to boost, but. I think the way it's described is very important, especially for children. So it's very important.

Daniel

Awesome stuff. Thank you all so very much for being on Foundational Impact. 

S3CURA

S3CURA

We are a team Secura at HU Prize and we basically help the security organizations to enhance their current CCDV infrastructure by giving them alerts for situations, situations which would have otherwise gone unnoticed. So, for example, right now, these systems can detect objects, for example, a human being or a car or a gun in rare cases, but then a police officer might have a gun and then that is not a, that serious of an incident, right?

So now the capabilities exist, uh, because of the innovations that has been done in the last three, four years, 2, 3, 4 years, that you can understand the nuances of these situations, just like a human being would. And you can give really specific alerts and which really cuts down on the false alarms because you can imagine if you're just raising an alert based on human detected, then 99 of the a hundred times it would be a false alarm, right?

We are personally from Chicago, a place with historically high crime rate. And this is an issue that we feel technology can address. Technology can help solve it. For example, there are to any kind, there are two aspects. One is reporting and the other one is response. The response is in the hand of the police officer or the operator, but the reporting is the hands of the citizen who generally notice, and sometimes people are scared to report or it takes a lot of time to report.

And we thought, what if we can transform cameras into witnesses? What if cameras can report by themselves? And that's how we started developing the AI.

Daniel

Surveillance is a really interesting one. Um, so I, um, obviously from the education space, we speak to a lot of schools mm-hmm. Around things like data privacy, intellectual property, and various other things. If you, if you look at that from the perspective of surveillance, there are a lot of, I guess, ethical questions around what we're surveying, how that's being surveyed, who has access to the data and information.

Can you talk to us a bit about how you're addressing that or how you're thinking that through with your business?

S3CURA

Sure. So what we essentially do is when sending data, most of these people can, it can be accessible to the internet, right? But when we send data over to our cloud, we encrypt the entire pipeline.

We establish a unique VPN between where the cameras are and the cloud, so that data pipelines could not be hacked in the middle. The second thing is a human operator or person watching the video can have bias and could be a racist person. But we encrypt the data, the facial data. So essentially you cannot make, the model that we have does not make decision on the basis of race, color, ethnicity.

So we encrypt the facial data and then a, an officer or a operator can decrypt that data when they want to investigate, but that is maintained as a permanent log. So that increases accountability only the organization side. And there we. Are, uh, going to have A-G-D-P-R compliance. Uh, so we are seeking two certificates called, uh, uh, SOC two, type two, and, uh, uh, eye surgery only 7,001.

So that helps with, uh, all this as well. Good luck with to you on both of those. Yeah. Significant undertaking undertakings. I, I want to mention though, adding to him that. Um, it's not just that we don't, uh, we have like trained the system to not make any racial profiling. It can't make a racial profiling because the facial data itself is not available to it to make any judgements on that basis.

The actions might be disrespectful, not the color of the skin or something like that. Right? So that really ensures like we are on the right trend. And to be very honest, we want to set a trend that this is the standard that these technologies should be at. So that even it is a forcing factor for other companies to, uh, maintain that standard.

Daniel

So thinking about these, um, these experiences that you're having with the technology, the requirements that you have of new employees, what might you say to school leaders in K 12 right now who are thinking about the kinds of skills that young people might need in the future when it comes to AI?

S3CURA

Mm-hmm. So, uh, one of the primary skills is critical thinking at all points of time. Uh, people should look AI at AI as an accessible resource. For example, if you have a thought process, now you can use AI to test that thought process. If a child thinks about, okay, how would, uh, mark Twain write in 2025 noticing about the modern scenario and vehicles?

He could just ask that and, you know, it would be able to create unique references. Yeah, so now you have a platform where you can, in a closed space, test your creativity because. It is, it was so difficult. I remember when I was in school, we had to browse through Wikipedia to get important points, but now I can get access to information much [00:12:00] faster and apply my creativity on it.

Yeah. So now, uh, growing up as a child has a hundred different questions. Now those questions can be answered from a credible source. Yeah. So I think that is one big thing. I don't know, like when I opened Wikipedia the last time, uh, I, I would like to, uh, add one more point that. In this era where you can get answers like really well from an ai, what is important is, and what nobody else can tell you beside yourself is what do you want and what do you want to ask?

The question is, what questions to ask is more important than the answer itself because you can get the answer. It's like you have an Oracle and it can give you any answer, but what will you ask? What is the craziest question or. And you have to understand its capability right now and ask it in that way so that it is most beneficial for you right now.

And I think, uh, teachers can, uh, be taught in that regard that as to how to get the most out of these, uh, models. And then, um, yeah, even like the students, how they can like learn about prompt engineering and so on. 

Daniel

Excellent stuff. What a way to wrap up. Thank you so, so very much indeed. Thank you so much for your time today. Really, really appreciate it.

‍

Cocoa Potash

Cocoa Potash

I come from Ghana and I'm with Coco Potash. We are recycling agro waste, um, cocoa port husk into potassium carbonate. 

Daniel

Wow. Can you tell us about why this is a significant project for. For you, you said you're in, in Kumasi, in, in Ghana. Can you give us some context there?

Cocoa Potash

Yeah, so, um, Ghana actually is the second largest cold co-producers company, um, country in, in the world. And the whole Kuma, 75% is the husk, but mostly it's strong away waste and most farmers try to use, uh, unconventional methods to get rid of. Actually it serve as a substrate for the fungi that caused the blackboard disease to the cocoa farm.

So I come from a cocoa farming community. My family also relied on cocoa farming. And my dad in an attempt to get rid of the ports, actually set fire into our entire 15 acres, acres farm, and we lost our livelihood because of that. So when I had a chance to study chemical engineering, I decided to find a solution to the waste. And that is where I learned about its, I mean, content of the potassium carbonate. And we began using our own technology to transform them into potassium carbonate.

Daniel

Can you tell me about anything you might be doing from a, a technical perspective?I'm obviously very interested in how AI might come into this. Um, anything you might wanna share on that side? 

Cocoa Potash

Yeah, so on the technical side, we want to be able to track our day-to-day impact, both on the environment and on moment. And so we are looking at a dashboard that can actually give us a day-to-day kind of tracking of how much waste we are taking from the environment, our footprint as a manufacturing company and how that is balancing out to give us a net positive impact on the environment. That is where we will need to explore AIi. And we are also looking at how we can also use AI in our, um. Um, talent acquisition and maintenance in the company. 

Daniel

So what might you say to a school leader who is thinking about banning or blocking the use of AI for students? 

Cocoa Potash

Well, AI has come to stay. That is the future. You cannot be blocking the future.I mean, if you really want your, your students to be, uh, part of the global market, even in the local context, AI has come to stay. Its responsible use is what we have to perceive.

Daniel

What a fantastic way to wrap up. Prince, thank you so very much for being on Foundational Impact. Welcome. It's wonderful to have you with us. Thank you.

Sara Technology

Sara Technology

Um, so we are building AI speech tutor for children with speech difficulties or, uh, language learners, um, that's struggling with pronunciation. So we provide them with personalized speech exercises so then it can improve the clarity of their speech and increase their confidence.

Daniel

Can you tell us a bit about, without going into Yeah. The, the secret source? Mm-hmm. A bit about the backend and how, how you're building this and what, um, what's going on behind the scenes. 

Sara Technology

So basically we're trying to personalize, um, um, based on the user or the kids' performance, like how the kids is, um, having issues with their speech difficulties. We're trying to help them based on their issues within their difficul. Um, and with that we use a little bit on AI based on like how we detect their issues with their, like speech. How do we like, get them the right content, the the right, um, advice to help them improve their speech. So everything there is personalized.

And for the personalization we use AI. Okay, so you're using APIs. Um, yeah. Yeah. We use, um, like remote, the backend. We, of course, the AI is in the backend, in the cloud is like engine, um, and trying to understand kids, uh, speech difficulties, everything else. Yeah. Um, and we're also using like pregenerated, uh, content using AI, but review backend. Okay. The main AI application here is that, um, we are detecting the children's. Speech at the sound level.

If we hear them saying things right, we just keep encouraging them positively. But if they, we hear them saying things wrong, then again, we provide these kind of gentle cues. Um, and, and why I say gentle cues is because we want, we don't want to, there's two reasons.

Um, one is on responsible AI. There's no model out there right now, at least to our knowledge that is a hundred percent accurate. So we want to make sure, um, we are not telling the child that they are absolutely right, absolutely wrong when the model is not a hundred percent accurate. So while we design it, it's more, um, maybe the model is like around 80% accurate.

If they said it wrong and we think they said it right, then we just say, great efforts, carry on. Okay. If they said it. And we say, think they said it wrong, then we give them the cue. So it's, it's not a harmful process, but generally at the time, most of the time it's accurate.

The personalization algorithm. We decide what is the next practice they would go to, what is the next game they would go to. They are actually AI based and we do this personalization using the AI. 

Daniel

So what might you say to school leaders who are thinking about. The skills that young people at school might right now might need when thinking about university and the workplace beyond. As far as AI is concerned, I. 

Sara Technology

Yeah, I think it's undeniable that this will continue to have these kind of huge impact, like in, in the, whether it's personal life or work life.

Like I use AI for all range of things, including becoming my therapist for personal life, but also like on the work side. Like for example, before Love Lovable came out, it's um, it's a website, web building tool, web building tool. Yeah, using AI before I came out, we actually spent, uh, quite like maybe a month or so, like building our website, you know, it's actually quite costly, like a thousand dollars or something.And I, we, I still don't love the website. And then Lovable came out and then I designed the entire website in like a few hours. And then the next day we do like, implement it also in a few hours. And then we have a brand new website that like looks 20 times better than what we had before. And I think that's what technology is able to do to increase our productivity. So I think all. The, the bar is so low to use these kind of AI tools these days, I feel like, um, all the schools should welcome it. It not, it, it, there's just no way around it. Um, and I, I don't work for Lovable, but I do love their product and there is a student discount. A great practical tip to wrap up with today.

Daniel

Thank you both so, so very much for being a part of Foundational Impact. It's been wonderful speaking with you.

‍

Carbon2Capital

Carbon Capital

Carbon Capital in a nutshell, is a sustainability FinTech platform that helps banks, businesses, and individuals to understand their carbon footprint and act on it. One framework that we looked at, and it's something we also pitch all the time, is we wanna make it as, when you look at the Myas or whatever and you know, the car, the calories of it, we kind of built on how calories were pushed onto us. Hmm. So now people are aware of calories and they know how much calories. So we want to do the same thing with carbon emission. So you look at an object and you know, the carbon emission or an estimate of the carbon emission

Daniel

You're using AI to do that? 

Carbon Capital

Yes. Yes. Yes. 

Daniel

Can you tell us a little bit about how that works. What you're building without telling us any secrets, of course, about the organization, but as much detail as you can around particularly the backend.

Carbon Capital

Yeah, so from day one, carbon to Capital, AI first company. So everything that we start working with, we think like, how can we incorporate AI because this is the future. Uh, so as Angelo was mentioning, like, uh, for each transactions that you make, like we give a give analogies and these analogies are understood, like, uh, if you just give numbers, the users doesn't understand that right.

So we use AI like heavily to explain the numbers to the users in simple language. So instead of saying that, okay, you have 16 kilograms of, uh, carbon emissions, we, we simplify that. We understand the AI understands the user's behavior and the things that the users can design it to. And based on that it gives like very specific examples or, uh, analogies that you would understand, you could relate to, for example, like if you are more into like, uh, let's say, uh, food. Yeah. Like if you, if you like food, like it'll give a, it'll understand from a transaction history that you purchase a lot of food. You go to a lot of restaurants. Mm-hmm. So it'll give some allergy that is related to food. So you as a person can understand that how much in fact you're living on the environment.

Daniel

The majority of our audience work in education, we have teachers and students, et cetera. I'm wondering if you had any advice for school leaders. Around the kinds of skills that young people might need if they're looking to the future of work. When taken into consideration the work that you are doing and the direction of travel for your, your company, particularly around AI, is there anything that you would wanna put more attention or focus on? 

Carbon Capital

I think, um, uh, recently a few of her assignments were based around using ai. So I think that would be one thing, like I think it's very important, um, in terms of from engineering and in terms of, uh, creative thinking as well, that when you get a task, you see which part of this task needs creativity and a human input, and which part of this task can be auto automated.

Because if it was not for us being able to have access. To tools like this, this, this project would've taken us years. Exactly. That's what we're saying. So I think it definitely helps. But what's important right now is that, um, in school. Users do not know how to use it optimally, which again adds to the environmental impact as well because they're doing way too many goals and using it for every small thing and then the output is not as good.

I think it's important that prompt engineering is something that is kind of is taught to the student at school because that quite, that helps a lot. And then. Yeah, also knowing where it'll actually help and where it's going to waste time and course as whether if it's going to have like a negative environmental impact.

Yeah. Also, like in terms of like, if I think from programming or developer perspective, I think this is the best time in the history where living, uh, to be like, uh, to create something or build something. So like at this age, like you don't really need to learn like the nitty gritty of all the programming languages send taxes, right?

If you can just think. Like plan something on your head and you can just think it out loud and transfer the knowledge to the AI model and optimize the, the out all the. Great tools that are out there, like you are good to go. Like, I think I'm, I'm pretty sure like in like the next years we will see like unicorns that are done by only one person.

So like, uh, we're going to one, uh, like a paradigm shift that is like crazy and this is the best time if either you learn the AI or, or like you would be falling behind significantly. Yeah. So it powers you. I think that would, that would be like, don't, don't be scared. The AI just. Use it correctly so then it could empower you way better. So yeah. 

Daniel

It's a fantastic way to wrap up the conversation folks. Thank you so, so very much. Thank you for being a part of the series.

ALBON

ALBON

We're a climate tech startup focusing on turning carbon dioxide and wastewater into a sustainable fertilizer. 

Daniel

Excellent stuff. 

The solution that you're focused on. So if you are a, a company I suppose that's trying to address waste management, how did they get in touch with you? What are you able to offer them? I'm just trying to get an understanding and a bit of context for

‍

ALBON

Yeah, yeah. I guess I can give a little bit more of a top down of the company. So. When an abattoir slaughterhouse rendering facility dairy farm releases their wastewater, it's full of essential nutrients that typically trees love to grow in.

However, the problem is this water is just ending up in the natural environment. Mm-hmm. Where we see toxic algal blooms take place with a hold of that nutrient. Now the problem is there's a little bit of a disconnect in our cycle in how we get those nutrients back into our soil. So what we do is we have created a system and we use a good strain of algae to pull those nutrients up from the bad water and then turn that into the fertilizer to return it to the soil instead of just flushing it down the drain.

Daniel

Can you tell me about your tech setup and infrastructure for the organization?

ALBON

So we have a prototype, uh, unpaid pilot going at Wiley Park. Um, see our PBR system. Um, right now it's just being operated by some mortar farms and couple raspberry PIs. Very low budget. And we are looking into integrating our AI software.

To operate the, um, photo bioreactor, uh, farming market argue is very much like any sort of farming process. It's not quite chemistry, as in it's, uh, an equation that you can just plug in. So we need to develop a sort of farm's intuition in developing our air model to learn to grow and harvest out of algae.

We're working at a local park in Sydney at the moment. Um, we got in contact with them through just showing up to local river cleanup events and talking to the community leaders there.And they had identified a local pond that was having pretty bad seasonal algal blooms that was resulting in dead fish. Dead birds didn't smell very nice. Um, so. We thought that this was a perfect test case, um, for, to see if we could slaughter solution in. And like Justin said, a lot of it has been sort of feeling it out, figuring out how we can best implement AI into this wicked problem and sort of make it something that's scalable and that we can spread across, not just Australia, but hopefully across the world as well.

Daniel

At the moment, how are you looking for graduates or or university students to become part of your organization who have a degree of AI literacy? Or is that part of your recruitment process? 

ALBON

Are you asking them about their use of AI on a day to day, or not yet? I think, um, it spans more than just the actual development of the ai. I think when it comes to AI literacy, um, being in a startup space, a big hallmark in any entrepreneur will agrees the various hats you have to wear, because as CEO is not just direction of the company, it's also how to manage the people on the ground, how to framework policy into what we're doing.

Even the marketing side of it is all sort of wrapped up into one role. [00:10:00] And the same can apply for anybody else on the team. If Justin needs to focus on something on the biology side as well, we can collaborate. But if we're not always able to at that moment, that's when AI can come in handy and help push that needle along towards, um, and improve our development cycles.

Daniel

I'm obviously interested in the school space, right? Mm-hmm. This is what Good Future Foundation spends a lot of time working on. What might you say to school leaders or educators. Who are thinking about the skills that the next generation might need for the world of work when it comes to, to ai? Is there anything you'd want them to focus on?

ALBON

I think, um, particularly with Generat ai, I think the focus should be asking the right questions. Hmm. If you ask AI the right question precisely, it can generate you a good output. Just like if you, if I were to ask you a question. If I need, I need to frame that question precisely in order to get the correct answer that I'm looking for, which is, uh, [00:11:00] I think it's an important skill, um, which with all, with ai it's quite an important skill, but with AI it will advance you like very far. So yeah, for critical thinking, yeah. 

Daniel

Excellent stuff. 

ALBON

It's an incredibly powerful tool. I mean. It's hard to, I guess, cater this to a generation when you have a big black box that gives you all the answers without having to even do the critical thinking yourself. But I think framing things as problems to solve with AI and, and I don't know, this might be easier said than done, but adding steps to it, um, having some back and forth with an AI and having it challenge you. I think is probably a good way to utilize it, to actually engage that critical thinking so you're not just getting the answer, whatever that might be. 

Daniel

Awesome stuff. 

‍