Meet the community: Nadia Piet
Member Highlight— Introducing the founder of AI x Design
_________________________________________________________________
PSA: It is a key value for us at AIxDESIGN to open-source our work and research. The forced paywalls here have led us to stop using Medium so while you can still read the article below, future writings & resources will be published on other platforms. Learn more at aixdesign.co or come hang with us on any of our other channels. Hope to see you there 👋
In between sipping oat lattes and walking through a Python script, we sat down with Nadia in her living room in Amsterdam for the first of our Member Highlight. We caught up with what she’s up to, her journey into AI x Design, and her thoughts on AI products and capabilities. Read the interview below and tune in to our conversation.
Hi Nadia! What are you doing at the moment?
I’ve just quit as a design researcher at Bit and started my master’s in Data-Driven Design. Besides that, I am building the AI x Design community and working on some fun, creative, and personal projects (non-tech related). Most of the tech that I do is in school and AI x Design and everything else that I’m doing. So outside of that, I try to do some other things like collaging, making scrunchies, and writing.
Why are you drawn to AI x Design?
As AI is finding its way into every industry and area of life, we can no longer afford to excuse ourselves from engaging because it’s too technical. AI requires a multitude of people, perspectives, and practices
to shape its impact and assign meaning. Besides it being one of the most important challenges of our time, I also found it to be a field in which I can combine a lot of the things I’m excited about — design, tech, research, philosophy, cross-disciplinary collaboration, and accessibility and inclusion. I can do it all, instead of having to choose between them. It’s kind of a blank canvas.
What’s your background and how did it lead you to this — AI x Design? How did you get here?
I’ve been freelancing for seven years around design and strategy, exploring a lot of different roles and fields to see where I belong. Two years ago, I started doing more critical and creative research.
I’ve always been interested in tech and AI, but also a bit intimidated by it, like most designers. At some point, I had to do my thesis and I decided to learn more about AI. I figured that whatever I will go on to do, AI will be everywhere. I figured it’s going to be a good investment to understand it so that I can make more impact, in terms of having an opinion about how we design these systems.
I was also curious about what AI means for people like me (designers). What is it? How is it going to show up in our lives and in our practice and in our industry? So I started my thesis around that question and basically never came back from it.
That materialized into the AI meets Design Toolkit. And that was just scratching the surface. I realized that there’s so much work to be done in this field; work that is necessary and important. There’s a lot of really interesting topics to explore. It’s kind of a blank canvas almost. There’s not that much done around this topic, which makes it more interesting. It’s not about implementing best practices. We’re really asking new questions, and I think that’s why it’s really exciting. And I just haven’t gotten bored with it yet… so I think I’ll be on that for the next five to ten years at least, and maybe a lifetime.
What are some things you’re looking to learn more about?
I’m looking to learn more about Explainable AI — interpretability and explainability of models. In the first part, how do we, as in engineers and whoever’s managing the system, interpret its inputs or outputs? How do we assign meaning to the computational conclusions? And then how do we explain and help the user make sense of that output, so that we don’t just follow whatever it says.
Another thing I’m interested in is: How do we ensure inclusion and fairness? How do we do that as we go? Not afterward and be like, “Oh, whoopsie, we didn’t really have a representative dataset.” How can we integrate that as a default part of the pipeline and the process to ensure better systems? I’m interested in actionable ethics with a focus on inclusion and representation.
And I am learning to write Python! I want to learn TensorFlow and also learn how to properly use coding tools like Colab and Git. I want to be able to train my own GANs and GPT-2 and use small machine learning models in the browser.
Those are three things I’m learning more about, but I could definitely make a list of a hundred things.
What’s your favorite AI feature in a product/tool/service/interaction?
Some of my favorite ones that I use on a day-to-day basis is the translation feature on Google Lens, which is a combination of character recognition and machine translation. You can hover your camera with Lens over a piece of text and it will translate it in real-time on your screen to English. I found that a very helpful feature while traveling.
One of my other favorite functions is the object recognition in Google Photos. If I put ‘cat’ I get everything I’ve ever taken a photo of with cats. Often, I’m looking for a specific moment, thing, or object I’ve seen at some point in my life and I just put ‘yellow book’, for example, and it retrieves it. It’s like a little hack for my memory.
And then, of course, I absolutely love the Spotify curation and recommendation algorithms.
What are you most anxious or concerned about in the field of AI and Design?
That AI will increase, not close, the inequality that exists, and through a greater economical and digital divide. If only a very small portion of the world knows how or what AI is — how to use it, how to write it, how to interact with it — and only a very small portion of the world is represented in those systems and in the datasets that feed into it, and only a very small portion of the world is financially benefiting from those systems, while the others are being displaced and ignored because they’re not in the system or excluded because they don’t understand. That’s not something I would want for the world and it is definitely something I do see, unfortunately, happening at times.
I’m worried that it will benefit and represent those that have, historically, already had all the privilege. And the people who have been historically oppressed will now be algorithmically oppressed.
This is not based on malicious use. I don’t think people will use algorithms maliciously — maybe they will, but I think it’s more helpful to think they won’t — but that just because of our own biases, our own privilege, and our own very limited understanding of the world, we will accidentally put this algorithm out into the world and make this happen because we’re stupid, not because we’re mean. We’re human and we’re very flawed.
For AI to assume that whatever you are showing in your face and in your voice is the ground truth of how you are currently emotionally experiencing the world, that seems problematic.
On the brighter side, what are you most excited and hopeful about?
One thing I think is really interesting is that AI is genuinely able to find insights that we never could have. For example, with applications in healthcare, it’s able to analyze diagnostics with treatments, with genetic disposition, so that it can really understand what the best treatment for you is. In a way that a doctor could never do, because they cannot process all this amount of data of all the treatments ever digitally documented. And an AI can, so it can find quite novel patterns that might be really important. I think there is real value to be found in very advanced data science in understanding ourselves and the world around us. So that’s something I’m quite excited about learning; using it as a tool for learning and scientific investigation.
The other thing I’m excited about is just giving — we can give people access to things now at a scale which maybe wasn’t possible before AI. A friend of mine created this great app called Sis Joyce and it’s for people to talk to a chatbot to figure out if they’re ill. And obviously, this is not fully reliable, but these people might not always have access to traditional healthcare. I think because it can be largely automated, because it can scale very well, because it’s not constrained to a large physical presence, AI can give people access to things that previously they may not have had physical, financial, or cultural access to.
It can also help us optimize things like energy use and figure out solutions that we cannot see because we’re emotional and flawed. We can only hold a few ideas in our head at the same time, which is why we can do other things better, but the computer can definitely do some things better.
What do you think about Emotional AI though?
It is interesting, and I’m sure there is good stuff you can do with it, but it’s something I’m definitely more concerned than optimistic about. There are so many problematic things about it that I’m not sure it’s possible to design your way out of.
Why would my technology need to know if I am sad? If I am sad, I need humans, not models.
Who needs to know how you’re feeling? We know that you don’t always show your feelings on the outside — in your voice and in your face. For AI to assume that whatever you are showing in your face and in your voice is the ground truth of how you are currently emotionally experiencing the world, that seems problematic.
The second problem is: Who has access to that? For what reason? Do they need to know how I’m feeling? There are some good applications in driving cars (human-driven cars) where it sees if you are falling asleep. Now, that’s good. But yeah, why would my technology need to know if I am sad? If I am sad, I need humans, not models.
So, Emotional AI is something I’m a bit skeptical about, and ironically, one of the fields that are quite developed at the moment. It’s quite accurate across cultures as well, which is also problematic. For example, if you’re culturally mixed in a way that you look Asian, and maybe you grew up with more Western values [and inherit such gestures], what does that mean? Should we put you in the Asian model or in the Nordic European model?
So, a more fun question not related to AI: If you could bring one book to a deserted island, which would it be?
I would bring The Wisdom of Insecurity by Alan Watts.
I find him really great in terms of making sense of the world with a sense of humor. He has a really great way of reframing challenges in life in a very constructive but not overly optimistic way.
What is your favorite podcast (not necessarily AI x Design related)?
99% Invisible. It’s about the 99% of the world that is invisible to you — they’re basically stories around design and the world. Definitely design-focused with a storytelling approach and on a very wide variety of topics. Mostly things you have never heard or thought about. I find it really fun and inspirational. You really remember them, because they are told in stories and they’re usually quite surprising.
And if you had to pick a favorite podcast for AIxDesign related content?
From there on, our interview drifted into casual conversations and eventually into a work session as we cooked up some plans for our community.
To keep up with Nadia, follow her at @nadiapiet on Instagram or LinkedIn or get in touch with her at hello@nadiapiet.com.
Thanks to Kwan Suppaiboonsuk for conducting this interview.
About AIxDesign
AIxDesign is a place to unite practitioners and evolve practices at the intersection of AI/ML and design. We are currently organizing monthly virtual events, sharing content, exploring collaborative projects, and developing fruitful partnerships.
To stay in the loop, follow us on Instagram, Linkedin, or subscribe to our monthly newsletter to capture it all in your inbox. You can now also find us at aixdesign.co.
