Hello and welcome to the NVIDIA AI Podcast.
I'm your host, Noah Kravitz. Quick reminder, if you like the podcast, tell your friends subscribe and rate and review.
You can listen to the AI podcast basically anywhere you get your podcasts.
Apple, Google, CastBox, SoundCloud, and of course on the homepage for the pod on the NVIDIA blog.
And if you're subscribed, take a minute, drop us a review.
Five stars are always great, but whatever you feel and let us know what you think.
And you can contact us directly via email as well, AIpodcast at nvidia.com.
Drop us a line. Let us know what you think.
And also, if you have suggestions for topics to explore or guests to interview on future episodes, we'd love to hear from you.
Now then, let's get to today's guest, a very special guest.
Make it all about me, why don't I? I used to be a teacher way back in the day.
I have two kids in school and my wife is a career educator.
And so I'm very excited to talk to today's guest.
Tara Czachlowski is CEO and founder of Iridescent, a global technology and engineering education nonprofit that provides access to hands-on learning opportunities. that prepare underrepresented children and adults for the future of work.
She's been called everything from the pioneer empowering the incredible tech girls of the future, to, and you can't make this stuff up, a CEO science superstar hero.
I love that one. Tara is here to talk about a bunch of things, including the UN's AI for Good Global Summit at the end of May, this May in Geneva.
And an event that hopefully this podcast will air in time to plug and maybe get some of you out in the Bay Area out there in person.
Or if this is airing after the fact, then, you know, hopefully Taro is a great event and folks came out anyway.
The championship, the AI World Championship, which is part of the AI Family Challenge, It's taking place in Santa Clara, California on May 18th.
So we're going to talk about that as well.
And I think that's enough out of me because you can speak to it better than I can.
Tara Czachlowski, thank you so much for taking the time to come on the AI podcast.
Thank you, Noah. It's been an honor. Like I was saying, I've been hearing these podcasts for a while, so now I'm on it.
That's really cool. Well, I couldn't be happier.
We're recording on a Friday. It's a great way to end the week.
Where would you like to begin? Do you want to talk about iridescent?
Do you want to talk about the AI for Good Global Summit?
Where should we start? I can talk a little bit about why we are doing what we are doing.
So I'm a former aerospace engineer from India and I started Iridescent 13 years ago as a nonprofit, primarily to connect engineers and scientists from industry, from research, and to have them go into their local communities and inspire children to tackle big problems.
Over the past 13 years, we've tackled different types of problems using different types of tools.
Early on, it was mobile apps, Now it's AI.
It's been really cool that we've been working with Nvidia for a while, and now we're finally at the point where we can actually put everything around AI-based technologies.
And so we launched the AI Family Challenge last year, And we actually did not know whether people would be interested, whether there would be desire to engage in a multi-week program and whether there was global interest.
And so We created these hands-on design challenges with Nvidia and with some other organizations and we launched this competition to families saying, as a family together, find a problem that you care about in your community and learn create an AI-based prototype and actually try to build it.
We were blown away. We had about 7,500 children and parents from 70 different communities across 13 countries, and countries that we would not normally expect people to be interested in AI actually.
All of these groups spent 15 weeks really learning more about how this technology is shaping the world. and thinking about how they can contribute to the conversation.
And so we came away from that experience just very, very inspired by human curiosity, right?
And also, we have such an innate desire to learn and to feel part of a movement.
That's what we're coming off of, and now we're going to be celebrating the finalists. from Bolivia, Pakistan, Palestine, Uzbekistan, Spain, and the US Saturday on May 18th.
So it's been a very, very exciting year.
That's amazing. 7,500 people to sign up for a 15-week program.
In anything. It's incredible. In anything, right?
Parents and children. So I think... Everybody, I mean, from the day I started your desk and people would say, every school would say, oh, our parents are not engaged.
Our parents are too busy. Parents are not going to come.
Sure. How is the program delivered? It's delivered after school in the evenings, and I think the magic solution for us was to host it during dinner time. and to basically provide a little bit of funding for the school to cover the cost of the materials and the food.
But what families realized was everybody needs to eat dinner.
And when you open up that time, And then you have the family just spend time playing and tinkering and exploring and learning together and learning about one another.
Very rarely, families, regardless of the socioeconomic status, families don't get that type of time.
You're very used to dropping your kid off to a soccer practice or this after school activity, but it's rare for the child to see their parent learning something similar. and for the parent to sort of see the child reacting and sort of contributing to a conversation.
And so once the families come for that first session, They always come back because it is such a special bonding time.
And what we found is like sometimes when you're stressed, learning something new reduces your stress. especially when you're doing it with your loved ones.
And so that has been the core element And yes, the thing that really blew us away was the depth of curiosity people have about AI because it's everywhere in the media.
They hear about it. But every community came back to us and said, nobody bothered to explain to us what this was.
So this was the first time a program came to their communities and was fun and accessible and something that the family could engage.
So they did. What kinds of problems did the families try to address and what sort of tools were used?
What kind of solutions? Yeah, so I think my team spent quite a bit of time scanning the horizon for some tools that were accessible and It's still a nascent field right now, but we ended up using a platform called Machine Learning for Kids which plugs into Scratch, which is a visual block-based language.
And it actually was created by an IBM engineer for his daughter in his free time because he felt that There was nothing that she could connect with.
And now people all over the world use this platform.
So it runs on IBM's Watson. And It kind of presents like toy problems or examples that are very concrete. that show you how you train the model to recognize different types of images, different types of emotion in text,
So that was the platform that all of the family used.
The problems ranged pretty dramatically depending on the experience, which is why this program is powerful.
Some of the finalist prototypes are, there's one from Palestine where this girl created an image recognition system that was trained to recognize emotion in children's drawings and she worked with a psychologists to basically figure out what the threshold was and what to look for in an image. and to try to identify signs of bullying or domestic violence.
And they were actually able to reliably train it.
So all of these are not like market-ready solutions, but the whole point is that it's empowering a child and a family to think about that they have agency and that they have a role to play in the world.
So another one is from a family in Bolivia.
They created an image recognition system to recognize weeds in Lake Titicaca, which is their biggest lake in La Paz.
And it actually recognizes the weed. And then they've made like a little vacuum that sucks up that invasive weed and crushes it and gives it to animals.
Amazing. That is fantastic. So they actually built a physical vacuum.
Yes, you can see the video. It's like a little vacuum in a boat.
Oh, that's fantastic. And it runs on a raspberry pie and...
So, I mean, there are many elements to it that still, of course, like this is a prototype.
They only spend 15 weeks from scratch not knowing anything about it.
But what we were ecstatic about was firstly, this interest in doing something for such long period of time and then that people really need this kind of program because there's the desire to learn.
As we record this, it's a week from tomorrow, as we record this, May 18th, the World Championship in Santa Clara.
And so you called it a celebration, which is great.
Are the prototypes going to be on display?
Are they going to be judged? What's going to go down on the 18th?
Yeah, so the families basically get to pitch their problem, their solution, and there is a panel of judges that will be reviewing and they have a rubric based on sort of the depth of the innovation, the clarity of defining the problem, whether the solution addresses the problem and the family's own learning.
And then, yeah, all of these families basically get recognized because They are the ones that have received the top scores.
But yeah, so I think it would be a pretty formal pitch presentation and judging at the World Championship. really celebrating this desire to learn as a family.
Lifelong learning both for the adults as well as the children because In our work as an educational organization, we have another program that primarily targets middle and high school girls.
And it's a powerful program. But what we've found is that you need to put some pretty deep stakes into the ground in the community, and the way to do that is through the parents.
And the family unit, if the family unit is strong and lifelong learning is a core value, then the community will be resilient.
I mentioned at the top, I'm a dad, you know, within the past 10 years myself.
And the idea of doing that with my children, with my family just sounds so amazing.
So that's fantastic. And it brings the family much closer together because you rarely work on a problem as peers.
You have to manage the project together.
You have to understand each other. All of the team dynamics are reversed and they're all muddled up.
Because your child is curious and probably much braver than you are.
And so all of these things are just very, very interesting and it really helps bring the family closer together.
So are you going to run the program again next year?
Oh, yeah. Absolutely. Great. That's great.
I was going to mention at the top, and I didn't want to interrupt what you were saying, but sort of half disclosure, half... sort of getting more into what Iridescent does.
You recently received a grant from NVIDIA. focused on AI literacy and skills-based volunteer opportunities.
Yep. And so is that going towards funding the AI Family Challenge or other work that you're doing?
Yeah, definitely. A big finding that came out of our first year was that there is a need for technical knowledge to be spread out throughout this network.
And it's very challenging when in many of these areas that are very, very low human development index.
They have low social capital, social networks.
So we have to figure out a way to get technical expertise into these communities. supporting the educators that are running these programs and so mentors from industry will play a very big role and that was a part that was missing last year just because we didn't have the the bandwidth to include that.
This year, our goal is to launch the world's largest AI mentoring program.
NVIDIA is funding it, and it's based on our years of engaging employees from industry, It's very, very powerful because for the employee it's a way to, and they don't need it's a great way for them to be exposed to sort of the basic concepts and fundamentals of AI, which they probably hear about all the time in their work, but are too embarrassed to say, look, I don't know.
So this would be an official way to say, okay, we are going to become mentors.
And as part of their mentor training process, We will go through these fundamentals and in a fun way, right?
Who wants to go through sort of a boring online course that marches you through different slides?
If you do this in a collaborative way, in a project-based way, which our curriculum provides, And then you get to test it out and mentor a local family in your community or even your own family.
I mean, that really cements learning. Right.
So I think that's our pitch to different industry partners to say look most organizations like when we had the mobile movement they were moving.
There was just a one-side mobile department, and now every unit in an industry has their own sort of mobile app. piece to it and I think AI is going to be the same where currently it's sort of on the side but it will become integrated into everywhere so everybody will need to understand how to incorporate and leverage AI technologies.
So that kind of literacy is needed and we can be ahead of it or we can be playing catch up.
That's why I want to do this using the best practices of teaching.
None of the corporate professional development training methods are based on best practices of pedagogy, right?
That project-based is not part of it. It's not fun.
It's not hands-on. And so that's what we want to bring to it.
Our guest today is Tara Czachlowski. She's CEO and founder of a global technology and engineering education nonprofit called Iridescent.
And we've been talking about something called the AI Family Challenge, which is getting Families, not just the kids, but the families all around the world solving problems with AI, learning AI skills, applying them to solve problems. problems.
And this year's program is culminating in Santa Clara, California, in Silicon Valley on May 18th with the AI World Championship.
Let's switch gears just a little bit and talk about what you're going to be doing after the 18th. which is heading out to Geneva, Switzerland for the UN's AI for Good Global Summit.
And you're gonna be leading an education track, is that right?
Yeah. The UN and the ITU and XPRIZE, they launched the AI for Good Summit, I think three years ago.
This is the third instance. They asked me to chair the education track based on the work we did with the AI Family Challenge.
They asked me to think about what were some ways in which the field could really leap ahead and make some progress. that would go beyond what iridescent does so that was a hard challenge and i spent some time thinking about it and and i came up with five directions and one of them is the AI education, awareness in your community,
And the second one piece is the AI literacy and mentoring in the workforce that is building off of the grant that NVIDIA has given.
And then there... three other tracks that I think are important to think about.
So one is, demystifying AI through the media where we are working with journalists around the world to try to understand, especially those in small cities, and local newsrooms to understand what are their needs, what are their, and to provide some sort of toolkit for how they can get up to speed with AI because things change so rapidly, and then also connect them with basically like a think tank or a group of AI experts on different topics from all around the world.
So if they quickly wanted to get like a crash course on autonomous cars or NLP, right?
Like there would be somebody in their continent, maybe.
Yeah. And then another one is really sort of understanding the impact of AI-based games and apps on our mental health, especially children.
There's a lot of misconceptions around screen time, especially in Silicon Valley.
There's this thing like, oh my God, you can't do more than one and a half hours.
All of that actually, the research doesn't back that up, interestingly.
Those are some conversations that I'm bringing together.
Ideally, there would be a project that's launched that is collecting some data but also very definitively seeing what is the impact on a family as a whole.
So how can these kinds of technologies help families come closer together?
So one recommendation recommendation is don't fight your kid playing video games, but play with her.
And then the last one is there's a lot of talk around AI-based tools in learning and teaching, but In my mind, they're very narrow and myopic where they're teaching to some sort of test, which is kind of arbitrary.
Honestly, the future skills are really about problem-solving and creativity. persistence and so one question that's more interesting to me is how can AI-based tools really deepens student-driven learning and self-interest.
So many of these platforms, children would not touch with a barge pole if school did not tell them to.
Uh-huh, right. And so can these ed tech companies come up with platforms that really are helping develop problem solvers and creative thinkers?
There's a ton there that you just spoke to that, you know, you spoke to it in about, two minutes and we could spend about two weeks diving into each of these topics separately.
But I'm going to hone in on the last thing you said.
When you talk about AI enhanced, let's say, platforms and teaching tools.
What does AI bring to the table that otherwise wouldn't be available in a learning setting?
Right, and AI itself is such a broad term.
Exactly. Right? So I think if we were to strip away the AI term and just talk about data science, right?
I think there's a lot of talk around personalized learning, and I think it's very limited.
My point of view there would be, ask this question of, what role does this AI-based tool play in conjunction with the parent, with the teacher, with the peer, And if you were to map interest over, say, a period of five years, what role does the AI-based platform play in the ebb and flow of interest?
When you ask that question, then you can begin to back out what types of data should we be collecting that will help us shed light on this question.
I think a lot of people think, oh, we're just going to put like smarts into it and NLP into it and somehow magically it'll become personalized.
But I think a lot of it depends on your depth of understanding of the problem and if you don't phrase your question correctly or courageously enough.
I feel like too many tech companies are just asking very small, narrow questions.
How can we improve math performance in this test and can we empower students to be creative thinkers and problem solvers?
Then yes, an AI-based platform has the power to collect a ton of data, But then you need to be telling it what data is important and what insights need to be gathered so you can provide that.
Right. When the topic of technology and education, let alone AI, comes up, I immediately think of the primer example. in Neal Stephenson's novel, The Diamond Age, right?
And that's been the holy grail for years and years and years in some people's heads and the one that I always think of.
And that's a ways off. But this notion of personalization is one thing.
But then to your point, kind of honing in on that, what does that mean and leveraging that towards...
Well, let's not just go towards getting questions right on a test.
Let's go towards this exploration of... critical thinking and problem solving.
And it's so huge. Let's just shift gears for a second here and go into your background and your past.
You're in the Bay Area now, Northern California.
Did you grow up in India? Yes, 22 years there.
Okay. And so take us from that through to now.
Did you... Were you interested in computers?
Were you interested in family dynamics? Were you interested in education?
What's the path that led you to iridescent?
Right. Yeah, it's just an interesting question.
I've not been asked that about, did I have those interests?
And the answer is no, I grew up in somewhat like low income countries. lower middle class family.
And my father was a pilot and my mother was a doctor and both of them slightly unusual. in sort of the general sense.
And I grew up very differently from other girls because I was given a lot of freedom.
And I say this, I was brought up as a boy.
So I used to fixed engines and I used to mess around with cars and I used to ride motorbikes.
All of this is highly, highly unusual, right?
Even I guess in the US. What it taught me was that, honestly, to me, there's no such thing as intrinsic interest or motivation.
Do you think you'll be successful at something?
And usually you're successful at something if you have a lot of exposure to it.
And so I had a lot of exposure to mechanical things.
And so that was sort of my comfort zone.
And, and so I grew up wanting to be a pilot and, but I was too short to be able to do that.
That's the truth. You were just literally too short to operate a plane.
Too short to join the Indian Air Force. And so I spent...
Two summers hanging upside down, trying to increase my legs, and that didn't work.
So yeah, so I think I did the next best thing, which was to get a degree in physics and went into aerospace engineering because I wanted to design beautiful airplanes.
And yeah, so I was working at a company called Air Environment, which is now public, and they made the first drones.
They're based in Simi Valley. I really wanted to work there and make bird inspired airplanes, like really beautiful.
They made the first human. At the time, there was the Gulf War, and they started to make these drones that really...
They call them UAVs at that time. So I was working on that project, but somehow I just didn't feel that that was what I really wanted to do.
And at that time, there wasn't a startup aerospace industry. and so I had to really sort of step back and question whether that was the path I didn't want to work for a big aerospace company there was only one I had to sort of dig deep to, because that was something I had wanted to do for years.
And I decided like, okay, now I really don't want it.
There isn't an interesting enough sort of career path here.
And so- So what do you do then? I made a spreadsheet of two columns and one I put like, what are some big problems in the world? and inequity and sort of lack of access, but also sort of a lot of strife, right?
And it all seemed to stem from a lack of education.
And then on the other side, I listed strengths and I had only one strength and I put in, I love motivating people to do hard things.
And so that's how I started Iridescent. And the idea was to tap into engineers really and scientists because not so that they change their career path, but more that they share their love and their curiosity for how the world works and how to make it better.
And to bring that sense of agency to communities that normally would not have that experience.
And so that's how I started Iridescent. And so that was 13 years ago?
Yes. How has iridescent grown in terms of the work that you do, but also the vision?
And what else besides the AI Family Challenge is in the next year, a couple of years for Iridescent and for you?
Yeah, I think I see us as not sort of a typical nonprofit that delivers programs, but more as a design-based research organization trying to understand this question of how do you empower groups that normally are underrepresented in technology, how do you empower them to become leaders and to build a sense of agency, but also to equip them with the skills they need to contribute.
And it's a very big question. And I think over the years, we have sort of been chipping away And so very concretely, one question is, how many hours should a program be to have resilient effect on a child's sense of agency?
Right. And so for a girl, say, for instance, we have come up with this idea of like a hundred hours. which is the other program is Technovation, girls have to find a problem in the community, create a mobile app, and launch a startup.
That takes about 100 hours of team-based work.
What we have found is that the experience is so powerful and so unique, And so different from anything that they normally experience that it has transformative effect on their psyche.
And also on the community, because it is so different from what they're normally used to seeing girls do. that the community walks away thinking, wow, we just sort of supported this amazing new leader, right?
And so you don't have to keep doing things for 10,000 hours, but sometimes you can have this magic combination of experiences that can have transformative impact.
Those are the types of things that we're trying to understand that What is the role of the parent, the mentor, the teacher, the AI-based tool?
What should they be learning? and then what is the long-term effect of this to help build this creative problem solver, and innovator.
Technology is the fastest way to change the world, and so that's why that has been a key part of our curriculum.
And AI is our curriculum for now. I think we will keep refining it.
The AI tools change every day. A huge challenge.
I think climate change is a big thing for us to be thinking about.
And I think... If we can empower some of our communities to be thinking about how they can leverage technology to address some of these issues locally, I think that is something that we have this vast network that we can leverage.
I think in the future, not too far off. I think synthetic biology is something that is going to impact all of us as well and in very, very personal ways, especially in low-income countries where you may not have access to a doctor.
So I think those are all things that we want to make sure as an educational organization, we are educating students people telling them about the pros and helping them to be analytical so that they know what role these technologies play, what role their decisions have,
And then I think our role is really to sort of look ahead maybe five or 10 years and to see what would be impactful and then to create the curriculum and to empower the communities.
But We are honestly just a curriculum provider and a platform.
But yeah, it is helpful to sort of look ahead and be very closely tied to the technology innovators in Silicon Valley.
So we know what's coming. Clearly, you have lots to do.
So I'm going to... I'm going to wrap up our conversation as much as I don't want to and let you get to it.
But for people who want to find out more about about all the things Iridescent's doing, maybe come out to see the AI championships, maybe support the work that you're doing.
Where can they go online? Let's start with what's the iridescent website?
It's iridescentlearning.org and iridescent as in I-R-I-D-E-S-C-E-N-T learning.org.
Okay. And are you on social media, other channels?
Yep, we are on all of them. So Twitter, LinkedIn, Facebook, Instagram.
And then? And I hear that you have a lead role, so to speak, in a documentary called Code Girl.
I mean, the girls are the lead roles there, but yes, I am in that documentary.
Is that available online if people want to watch it?
Yes, totally. I think it's available on Netflix, Amazon Video.
Fantastic. Tara, thank you so much for taking the time out.
At the same time, the amount of things that you have your hands in, but also the passion and energy that you have for them. just really comes through when you talk about it.
And like I said, we could turn this, you know, 25 minute podcast into a... a 25 episode series, no problem.
So we we'd love to have you come back on, you know, down the line and maybe by then the.
The AI podcast will become the synthetic biology podcast, but you'll already be a step ahead of us working at it.
So we can talk about that. Thanks Noah. This was very fun.
Best of luck with celebrating the Family Challenge, the AI for Good Summit in Geneva at the end of May, and everything else you're doing.
Yep, totally. Thank you.
