The Product Experience

AI ethics advice from former White House technologist - Kasia Chmielinski

• Mind the Product

Share episode

In this episode of the Product Experience Podcast, we speak with Kasia Chmielinski, co-founder of The Data Nutrition Project, who discusses their work on responsible AI, data quality, and the Data Nutrition Project. Kasia highlights the importance of balancing innovation with ethical considerations in product management, the challenges of working within large organizations like the UN, and the need for transparency in data usage.

Featured Links: Follow Kasia on LinkedIn | The Data Nutrition Project | 'What we learned at Pendomonium and #mtpcon 2024 Raleigh: Day 2' feature by Louron Pratt

Our Hosts
Lily Smith enjoys working as a consultant product manager with early-stage and growing startups and as a mentor to other product managers. She’s currently Chief Product Officer at BBC Maestro, and has spent 13 years in the tech industry working with startups in the SaaS and mobile space. She’s worked on a diverse range of products – leading the product teams through discovery, prototyping, testing and delivery. Lily also founded ProductTank Bristol and runs ProductCamp in Bristol and Bath.

Randy Silver is a Leadership & Product Coach and Consultant. He gets teams unstuck, helping you to supercharge your results. Randy's held interim CPO and Leadership roles at scale-ups and SMEs, advised start-ups, and been Head of Product at HSBC and Sainsbury’s. He participated in Silicon Valley Product Group’s Coaching the Coaches forum, and speaks frequently at conferences and events. You can join one of communities he runs for CPOs (CPO Circles), Product Managers (Product In the {A}ether) and Product Coaches. He’s the author of What Do We Do Now? A Product Manager’s Guide to Strategy in the Time of COVID-19. A recovering music journalist and editor, Randy also launched Amazon’s music stores in the US & UK.

Randy Silver:

Hey, it's the Product Experience Podcast, and I'm Randy and I'm Lily, and it's been a while since we did an intro together.

Lily Smith:

Yeah, we were lucky enough to join Pendemonium earlier this year emceeing the Mind the Product stage, while the other person interviewed people for the podcast, and it's been great to meet people face to face, but we've missed being together, so we promised to do more intros together.

Randy Silver:

We met a lot of amazing people in North Carolina, but today's guest might have been my favorite. Don't tell any of the others. They're an advisor to the UN, they've worked with the White House, they were part of MIT's Scratch which might be my favorite part of the whole thing and they also helped scale the US's COVID response.

Lily Smith:

Kasia Kermelinsky gave a great talk on stage, then sat down with Randy to dig even deeper on data nutrition. In this interview, let's get right to it. The Product Experience Podcast is brought to you by Mind, the Product part of the Pendo family. Every week we talk to inspiring product people from around the globe.

Randy Silver:

Visit mindtheproductcom to catch up on past episodes and discover free resources to help you with your product practice. Learn about Mind, the Product's conferences and their great training opportunities.

Lily Smith:

Create a free account to get product inspiration delivered weekly to your inbox. Mind, the Product supports over 200 product type meetups from New York to Barcelona. There's probably one near you.

Randy Silver:

Hey, we're here in Raleigh at Pandemonium and I'm here with Kasia Chmielinski and they just got off stage and gave this amazing talk and we were hanging out yesterday and had some really good things to talk about as well, and we're going to get really into all this stuff, but first, for anyone who isn't here and didn't get a chance to see you, can we just do a quick introduction? Tell us what are you doing these days and how did you get into this world in the first place?

Kasia Chmielinski :

Yeah, thanks for having me. It's great. The energy here is fantastic. For those who can't see us, we're surrounded by small, bright pink dinosaurs which I love.

Lily Smith:

It's a good vibe.

Kasia Chmielinski :

Yeah, so my name is Kasia. I am a technologist. I've been building products for 20 years which is a scary thing to say as a product manager, and I've worked in a number of different places. At this point, I'm really focused on responsible AI and data and ethics, and I do this as a consultant. So I work with a number of organizations, including the Diet Nations, and I also run a small nonprofit called the Data Nutrition Project. We build nutrition labels for data sets.

Randy Silver:

Which is fantastic, and we will definitely get into all of that. Let's start with the UN, though. Tell us a little bit about the work you do with them.

Kasia Chmielinski :

So the UN is massive. So, shorthand, I say I work with the UN, but there's actually so many different components and I'm still learning the system. I've been an advisor with them for about three years. I want to say two different parts of the UN, both of which are really thinking hard about data quality, data standards and then how to use data either internal UN data or external third-party data to build and deploy algorithmic systems that will benefit the sector. So what that really means is I'm a team of one, but I work with folks who are build teams, who are on the ground, and we work together to try to integrate some of the best practices from academia and industry into what they're doing at the United Nations.

Randy Silver:

For those of us who struggle in just you know, single corporate bureaucracies or working in our own civil services, we know how hard it can be to have influence at that kind of scale. And you're trying to do it on an even larger scale and more interconnected. How do you make it work?

Kasia Chmielinski :

Oh, it's definitely not just me and I would say, I'm still trying to figure out how to have the kind of impact I want to have. It always comes down for me, it comes down to people. It comes down to meeting the right people at the right time who are doing things inside the organization and borrowing heavily and being humbled by their deep knowledge of the system. So this happened as well. When I worked in the federal government, I realized that I don't speak federal government.

Randy Silver:

I don't speak UN.

Kasia Chmielinski :

I mean, it's truly another language and I don't think that I will ever become fluent. And so what I've done is I've tried to find native speakers and I go to them and I say, look, I don't understand what this directive means. Or can you boil this down for me into something that makes sense? Or what do you want me to do in this situation and that's how I actually end up having an impact is I find the amazing teams doing things. I convince them and charm them into letting me work with them, and then we build things together.

Randy Silver:

That is a really smart way of making it work. That's very cool. Okay, I wanted to talk about something that you mentioned in your talk. You were talking about responsible AI, or a responsible approach to AI, and you kind of put it straight out there that the way that we approach product management as a whole is kind of antithetical to being responsible, and that's because we work really hard to maximize what we consider value, so we're trying to go for the biggest ham and we're doing things which inherently leaves things missing. Can you talk a little bit more about this dilemma?

Kasia Chmielinski :

Yeah, I think it's kind of at the core of product management and the way we build things, and it's both the process that allows us to make small innovative changes quickly that then we can scale. So it's definitely beneficial to building innovative products, but it also gets in the way of being thoughtful and moving slowly and with intention, and I think that it really ends up being a question that PMs have to face is how do I balance these things, how do I balance these approaches? And so what I mean is that, as a product manager, you know, I have certain goals that I need to hit. I need to reach my OKRs. I have some you know indicators that I'm tracked against. I want to get my bonus, you know, or?

Kasia Chmielinski :

if it's a really small company. I need to make sure that we launch on time so that we get the next round of investment. I mean, these are really big problems, right, and they're non-trivial.

Randy Silver:

And it's not an academic exercise.

Kasia Chmielinski :

It's not academic and it's fun. It's super fun. I think PMs were kind of like MacGyvers we're just in there trying to make things work, you know, and that's okay, because I'll get to them one day, but the decisions they make in the very beginning they end up being the DNA of the product, and that means that all those other people that were left out of the initial set of users will always be secondary. You know, maybe eventually they'll be brought into the fold, but it's always going to be kind of updating or refactoring the product as it currently stood, and that's where a lot of the gaps emerge that end up being identified later as bias or discrimination, or even just you know folks that end up using your product off-label, right? They kind of take it, they do something else for it and they find a way to make it work, because people are geniuses, people are really smart, but it wasn't meant for them to use the product in that way, and that's because it wasn't designed for them, and that's the kind of an issue that I'm highlighting here.

Randy Silver:

So how do we deal with it? I mean again, in a perfect world, we would build things more deliberately, more carefully, we would take as much time as we need. We have enough of a problem with getting accessibility baked in at the beginning and doing it right, but we're always in a hurry to. Let's just get it out, let's build something that doesn't scale, then let's add on this other stuff on top of it to make sure it scales and we're hitting all the regulatory compliance and everything else that we need to do. How do we actually deal with this?

Kasia Chmielinski :

There are a number of approaches. I think if someone had solved this, I wouldn't be here. I wouldn't be here, I wouldn't be talking, I'd just be following what they say. And I've done it a number of ways, sometimes more successfully than others. I think we can. Obviously, if you're starting something from the beginning, you have a great opportunity to center marginalized users, move the center right and actually build for folks who are at the margins.

Kasia Chmielinski :

And you can say I'm going to actually build for folks that I would usually cut out of the first round because I'll actually end up building a better product for everyone if I build accessible first right, for example, and so you can do that if you have the luxury of starting from scratch. Even then you're still making choices. And so I think that, regardless of whether you're starting at the beginning or you're coming in later, a really important thing is to actually be monitoring and tracking how things are going and to make sure that you have real feedback loops with people, not just kind of check box. I asked somebody did some user testing. I built a product, I launched it.

Kasia Chmielinski :

Well, did you go back to them and show them what you ended up building? Did you get feedback on what you actually launched or did you just do it as part of the design thinking process? Right, and I think if we build in places where we can actually get meaningful feedback and track the use of these products and change them over time, we can start to address the gaps in retrospect. It's not perfect, but I think that that's another approach if you're kind of jumping in in the middle as opposed to starting a redesign.

Randy Silver:

Okay. So a lot of times we talk about I'm getting started with something and let's thin slice, let's find one use case that we're going to do, and usually we choose the easy one. And what you're advocating for is don't choose the most obvious easy one, choose something a little more on the edge, whether it's marginalized people or diversity, or whether it's special needs of some sort, special needs of some sort, because if you solve that and one of the classic examples there is, you know, the use of closed captioning and things has made video better for everybody not just for people who are hearing impaired.

Kasia Chmielinski :

Yep, exactly, and it's a trade-off again. So if you want to design for the most complicated accessibility case, you want to design for, you know, users who are using five different languages instead of just one, and all of a sudden you have to have a multilingual interface. I mean, these things can get more and more and more complicated. You have to decide, kind of where your line is going to be. To your point, you want to actually launch something.

Kasia Chmielinski :

I get that, but I think that there are ways that we've just done things and we assume that it's the right way to do it. We should start with a process of saying what are we actually trying to solve and are there ways that we can increase the footprint of who will be able to use this product and it'll make it easier for us down the line, right? So if I build something and it's just this is a dumb example but single tenant, right, so it's just for one user, but I know that I'm going to be onboarding multiple At some point, I'm going to have to make it multi-tenant. That's painful. Do you want to refactor later or do you want to make some really basic database choices in the beginning to make it easier down the road. These are the kinds of things I think that we can dial up and down when it comes to the types of trade-offs we make from the very beginning.

Randy Silver:

So being deliberate with your architecture around things of we may not be turning this on now. We know we're going to have to, so we're going to leave space for it, even if we're not doing it today.

Kasia Chmielinski :

Yes, where it makes sense.

Randy Silver:

Where it makes sense. Yeah, of course. Okay. And then you also talked a lot about when you're in these situations. You have to sometimes make some hard choices about how you're going to approach existing as a product person in a company that is trying to realize value at pace and be responsible. So one of the things that you said is one of the options we have is that we can refuse to build and sometimes redirect. Yeah, Tell us a little bit more about that.

Kasia Chmielinski :

It's kind of the product manager version of voting with your feet.

Randy Silver:

Yeah.

Kasia Chmielinski :

You know, you're presented with some kind of a product plan that perhaps you did not build. It comes as an edict from somebody up in an office somewhere and they say that shall build the following thing, and you go oh my God, that's terrible. For whatever reason, it's moral or otherwise, and one option that you always do have is to refuse to build it. Now, this can be a disastrous decision for, for example, your career. Right, Also, it's a privilege to be able to quit a job, because that assumes you can get another job. Right, Maybe you need a paycheck and all these things, and so those things are real. But there is the conceptual option, at least, of saying I will not build that thing.

Kasia Chmielinski :

A way to soften that sometimes is to say I hear you, that's really interesting, Let me collect some information or do some research. And then to go back and say the thing that you want to build is not the right answer. Assuming this is true, the right answer would be the following right, and you can actually redirect and say I won't do the thing. Essentially, you're saying I won't build your thing, but I'll build my thing and it will solve the same problem, assuming you agree with the problem, and that, I think, is one kind of initial approach that you can take. That is, I would say, it's a bold choice but it's definitely one that we have.

Randy Silver:

But we talk a lot about. You know, people are always coming and saying can I have this? And one of the core skills of a product manager that you have to learn early in your career is how to say no without using the word no, and I think that's essentially what you're saying.

Kasia Chmielinski :

Sometimes there is power in saying no, and I think that's essentially what you're saying. Sometimes there is power in saying no, yeah, like I won't do it, but that I think should be used sparingly, otherwise you get a reputation for being the one that yeah, you can only do that so many times.

Kasia Chmielinski :

Exactly, and so I do think you can take a stand sometimes and say I do not think this is the right thing to build, I will not build it. But again, that is very bold. I've done it a few times, but I wouldn't overuse it. Mostly it is this other thing you're talking about, which is let me show you what I would do. That gets to the same problem. Or let me explain to you why the thing that you want to build is maybe not the right answer right now. There's all kinds of linguistic and communications tactics you can take there to basically move them off of the path that you don't want to be on and move them onto the path that you do want to be on.

Randy Silver:

We did an episode a few months ago with a guy named Steve Hearsom who's an organizational design and a development consultant, and he was talking about. One of the real challenges for us is that in product, essentially you're middle management let's not beat around the bush but you're being asked to influence and act and work with people multiple levels above you in an organization and above your pay grade, and it is a real challenge.

Kasia Chmielinski :

Yeah, and not only that, but you're not actually in charge of anything. Yeah, but you're accountable for everything. You have no real power. You can't make anybody do anything really, but you need them all to do things right, and so I think that there's a lot of nuance in the role.

Randy Silver:

Yeah, if it was easy, everyone would do it right, okay, another thing that you identified as things that we can do is we can build better. What do you mean by that?

Kasia Chmielinski :

So for that one, what I mean to say is that let's say that you're in the context of organization or a product or a roadmap and you kind of disagree with some of the decisions that are being made. Well, you can decide how to build that thing in a way that's going to be less harmful or maybe more beneficial, right, and so it's kind of an extension of the redirect, but maybe a little bit less brutal. The redirect is they're going in one direction and you shift them to another direction. I think the build better is you see the product, you see what the gaps are and then you try to kind of shift so that you can fill those gaps and you can address those errors. You can buy yourself more time, you can get the resources you need, you can get the expertise you need, whatever it is to actually address those issues as you go. And I think you know whatever it is to actually address those issues as you go.

Kasia Chmielinski :

And I think in the world of AI, if we want to specifically talk about AI, there's a real opportunity here to not see AI as a product but rather as a process and, if you broaden the aperture, to think about AI as a process that involves use case selection, training, data selection, the training of the model, the development, the deployment, the updates and then eventually the decommission training of the model, the development, the deployment, the updates and then eventually the decommissioning of the model. And you think of this as many, many steps where things can go right and things can go wrong. Then you're not just at the end of the whole process saying, oh God, we have a problem, right. You're able to then jump in and build better by saying let's now identify issues earlier in the process, let's have ways to check the data, let's actually know what we're looking for, let's have a series of metrics that we're tracking over time.

Kasia Chmielinski :

When we launch something, let's assume that's the beginning of its life as opposed to the end of its life. Right, and now it's live. Now we have to train it, we have to make sure it's okay, it's out in the world, it's affecting the world, is it drifting? Are things?

Kasia Chmielinski :

I don't mean to antipomorphize but, you know and I think these are all now places where we can build better, if we have a more nuanced and a broader perspective of what the product is. It's more of a process than a product. So these are the ways I think that we can build better.

Randy Silver:

So when we are building without AI, we have developed lots of tools, very practical approaches to doing testing and QA along the way. We've got BDD, behavior-driven development, we've got test-driven development, we've got unit tests. We've got all kinds of other essentially tick-listy kind of things and processes. With AI it tends to be a black box without an audit trail. Is there anything practical if we're taking a responsible approach that we can do, as you said, if we're making a process of things we should be doing along the way?

Kasia Chmielinski :

Yeah Well, ai is a broad term. So when you say it's kind of black boxy, it depends on what kind of system, it depends on what algorithm, it depends on the deployment method and also how it's contained. So I agree, if you go and you get an AI system from a vendor, that's really black boxy because you might not even be, it might be in a cloud environment you might not even understand how it's working. If you're building something internally and you're using a statistical system, an AI system, machine learning system that's actually interpretable in some way, then there are definitely approaches that you can take to understand the inputs and the outputs, or at least the distributions of the inputs and the outputs, against some kind of benchmark or against some kind of test that you would expect over time. When you start moving into the kind of generative AI, probabilistic, you're building systems where you're, you know, cobbling a lot of things together.

Kasia Chmielinski :

I would say my best kind of suggestion right now is we need to be thinking about evaluations, we need to be thinking about red teaming and these kinds of about evaluations.

Kasia Chmielinski :

We need to be thinking about red teaming and these kinds of newer approaches to basically prodding and testing the models Another way to do it from or in addition to that makes a lot of sense to make sure that you're building in a componentized way so that you can isolate each component and test it for certain things. So, for example, if you're going to use an LLM for something, instead of having the LLM do the retrieval and all of the analysis and also the input and the output and send everything back to you, so it makes all these decisions internally, split it up into pieces so that you can actually test each piece for what. The one thing is that it's supposed to be doing, and that way you can kind of isolate problems at least at that level too. So as the technology is evolving and we're finding new ways to measure bias and issues with the outputs, you can at least limit the surface area or the footprint of each component and test them separately.

Randy Silver:

Product people. Are you ready? The word on the street is true. Mtp Con London is back in 2025. We're very excited for Mind the Product's return to the Barbican next March.

Lily Smith:

Whenever I hear people talking about the best product conferences, Mind the Product is always top of the list. If you've been before, you know what's in store. Oh, that rhymes. New insights, strategies, hands-on learnings from the absolute best in the field, plus great networking opportunities. And if you're joining us for the first time, I promise you won't be disappointed.

Randy Silver:

We've got one speaker already announced. That's Leah Tarrin, who you've heard on this very podcast. With more to come. From the likes of WhatsApp, the Financial Times and Google, you know real people working in the field who will share real actionable insights to level up your game as a product manager.

Lily Smith:

Whether you're coming to the Barbican in person on March 10th and 11th or tuning in digitally, join us and get inspired at MTP Con. London Tickets are on sale now. Check out mindtheproductcom forward slash MTP Con to find out more, or just click on events at the top of the page.

Randy Silver:

So you talked about it as a process and when you are being responsible with the process or deliberate with the process, anyway, you have the ability to control and put responsibility in at each of these steps. But a lot of the times now we've got people coming in saying can we just add AI to this? Or we've got third-party vendors coming in and people frustrated with how long it takes us to do things in product development. So they're just buying something in and I'll pick on marketing or sales, because, well, we always pick on marketing and sales, but they have vendors coming to them saying, yeah, we can do that for you and it's a shadow IT function that comes in. How do we add responsibility into our companies when we're competing with those pressures? Or is that just outside what we control? What we control and don't worry about the things that are outside our scope?

Kasia Chmielinski :

We should definitely worry about them, because we'll probably be responsible for the things that they do.

Kasia Chmielinski :

So, ultimately, if your liability is sitting with you I mean, if your customers are interacting with your system and your system as AI that you don't understand, your customers aren't going to yell at the third party. They're going to yell at you, right? Because? Yeah. So I think there should, first of all, be a kind of vested interest in understanding how this stuff works. I also am not against outsourcing things as a principle. You know, if your real value add is one kind of feature set and the other features you could get by, you know, contracting with some vendor who's going to use AI, like that's not necessarily a bad answer, but I do think that at this point, we should be talking about procurement and we should talk about the types of questions that you would expect a vendor to be able to answer. So, how does their system work, right? Well, I mean, first of all, do you even need AI to solve the thing that you're trying to solve? So is there a need for them? Okay, let's assume that you actually do want to work with them.

Kasia Chmielinski :

Great, what is the system trained on? How do they train it? How does it work? How are they measuring success? What's their ground truth that they're comparing it to to tell you whether or not their model is accurate? How do they plan to update and monitor this once it's in your environment? How often can you expect to be receiving updates from them? At what point would they consider decommissioning the program? Or would you consider decommissioning or turning off the contract and then making sure those things are really baked into the contracts and the procurement process so that there's some accountability there between you and the vendor? I think these are the kinds of things that we have to consider if we're going to be contracting with third parties.

Randy Silver:

Yeah, sounds outside the scope, unfortunately, of what most product managers have in their job.

Kasia Chmielinski :

I mean, a lot of things are outside the scope of what's there. What even is our job description? It's not very clear.

Randy Silver:

Fair enough, but yeah, as you were saying, systems thinking is incredibly important and the ability to discuss this with other people and bring it to light is still incredibly important regardless.

Kasia Chmielinski :

Yes, I mean, if you are a product manager whose function in the company might be replaced by some kind of outsourcing, I think you do have, then, a bit of a prerogative to highlight. Hey, before you decide to put me on a different project, let's make sure the thing that they're doing is actually similar to the problem that we're trying to solve and that we're doing this in a responsible way. I think you have a little bit of power there, but you know it's tough being a product manager. It really depends on the situation.

Randy Silver:

Not for the timid. Yes. No but being a PM probably is not for the timid, so you might want to find a different job. Okay, you also said there was another one about creating new solutions, and this is where your other project comes in. Really well, so talk about that a bit issues in the product development process.

Kasia Chmielinski :

You can refuse to build things right If someone shows you something that's morally contentious. You can redirect people, you can decide to stay within the organization and build better, but in some cases, the gap is not even attempting to be filled by anybody, and so your only option at that point is to create something new. And the good news is that, as product managers, I think we're really well positioned to build stuff. The good news is that, as product managers, I think we're really well positioned to build stuff, and we already understand that we need to have, you know, a multi-stakeholder, kind of a panel of folks and voices to help us build things. We know how engineers think and designers, and we can bring together, you know, the higher ups and the folks who have funding and all of this to basically become little mini CEOs of things, and so this is what I did with some folks in 2018. My team's going to be so upset I never remember.

Kasia Chmielinski :

But it was a while ago we all had a fellowship at MIT and Harvard. It was a joint fellowship called Assembly and we were charged with thinking about the ethics and the governance of AI. And they said you've got four months, here's a bunch of coffee and some cookies, and come up with something anything. So we were puzzling over what we thought the kind of problem space was, and at that time there were a bunch of news cycles about the problematic outcomes of AI. Bias and discrimination is kind of rising social consciousness around this and what we noticed as practitioners was that these media articles were really focused on the output. They're saying these systems are deployed, they're in production and they're hurting people. But all of us could kind of see that most of these issues were issues that were just carbon copies of issues in the training data and nobody was focusing on what about the data? Why is no one talking about the data? It's not public, it's not transparent, no one knows what's going into it. But then they're shocked when bad things come out the other side.

Kasia Chmielinski :

Is anybody looking at the data. And we started digging around and even from our own experience, we knew that there are no standardized data tools and no standardized data approaches when it comes to identifying a data set for use in AI. And we were again in an academic institution and we're drinking a lot of coffee, we're eating a lot of cookies, and there was a box of cookies and I don't remember who it was, and they turn it over and they were like there's a nutrition label on this box of cookies and before I eat the cookies, I can see what's inside. Shouldn't we like have the same thing for data? Just like something like that? And so we call ourselves the Data Nutrition Project.

Kasia Chmielinski :

We emerged from that fellowship. We make nutrition labels for data sets and the idea is that you have a standardized tool and design that allows you to understand what's in a data set before you use it. And for a practitioner who wants to do the right thing you know we're not certification body, you know we don't validate all the things but for somebody who wants to do the right thing, we give them an easy way to communicate what's in the data and then an easy way to digest what's in that data.

Randy Silver:

What kind of things do you have on the label?

Kasia Chmielinski :

There are a bunch of different components and we're actually in chats to see whether we need to expand this for different domains of data, like large language, model data sets, you know. But the kinds of things that we have are actually more qualitative, because it turns out that qualitative information is almost more important in this kind of a scenario than quantitative, quantitative stats. People will run a data scientist will run those quickly and say what's the distribution, show me this number of missing fields, blah, blah, blah. There's some kind of core set of things you're taught to do, but qualitatively, I want to understand how did you fill in that missing data? How much was missing? How did you clean it? Who paid for this?

Kasia Chmielinski :

All of these kinds of things you won't see in the data set. So we surface things like that. We also have a little module that explains to you what the intended use is. So this is what the data was intended to be used for. This is the way that it has been used. This is what you should not do with the data. These are known issues and areas for practitioners to surface whether the data set has undergone ethical assessment or technical assessment, these kinds of things.

Randy Silver:

And how are you seeing it being used?

Kasia Chmielinski :

There are a few prominent ways that people use the label. One is in kind of a industry data set context, which is honestly what we thought of initially, and we've partnered with organizations like the UN, microsoft Research, various places at Harvard to talk about data and talk about labeling that data and improving the metadata that already exists on those data sets, and that's really cool. So in some cases, folks have actually used our schema In other places. We've taken components of ours and added it to theirs and those are live and out there.

Kasia Chmielinski :

There's also an entirely second approach that we didn't really consider, which is that the label is very, very useful for education and for teaching and so there are a number of curricula out there that use the label to basically teach data science students how to think about data and how to build data sets and what a good data set would look like. And DNP Data Nutrition Project does a lot of kind of tours where we give lectures and we work with students and we kind of jump in give workshops, because it ends up being a really great teaching tool.

Randy Silver:

Fantastic, yeah, and have you seen it solve any problems? The difference in output and outcomes in this one.

Kasia Chmielinski :

Yeah.

Randy Silver:

Usage is an indicator that it's being used, but what for?

Kasia Chmielinski :

Well, here's an interesting one. One kind of thing that we didn't really expect was that the quality of data output goes up when someone knows they have to put a label on it, so that's kind of a secondary effect. We really built a label to make it easy to find data and use it, but we were working with a doctor at Morrill Sloan Kettering who builds data sets of skin imaging and is very aware of the potential bias in that and kind of the distribution of skin colors is just not very well representative, as we all know. And so she was talking to us and said since working with you guys and building these labels for these large skin imaging archives, the quality of my data sets has actually gone up, because in building the label, I realized that there are some things I should have done a long way, because you're asking me such questions about decisions I made.

Kasia Chmielinski :

I should have made different decisions. So the next time I make the data set this is like a data set they refresh once in a while the next time I make the data set, I make different decisions. I think that's the kind of impact that is really cool for us, because it's not just about bringing transparency to data sets, it's also about improving data set curation period.

Randy Silver:

And if people are interested in learning more about this and getting involved or applying it, where should just Google the data, nutrition or yeah, datanutritionorg, but if you Google data nutrition project, it would probably come up. Kasia, this has been fantastic. I have one last question. You closed today with talking a little bit about your mission, and a personal mission statement is something not everyone has, but I found it really inspiring. So tell us a little bit more about your mission.

Kasia Chmielinski :

Oh, I don't know which part you saw.

Randy Silver:

I don't know that. I have a very clear mission statement. Okay, the slide I saw, I'll read it directly. Yeah, there will always be dangerous gaps in technology that humans need to fill, but we always have opportunities to refuse, redirect, build better or create new solutions to make technology that works for everyone.

Kasia Chmielinski :

I mean, I agree with that, so I'm happy to I also wrote it.

Kasia Chmielinski :

No, I mean, I think that's it and it's maybe that feeling, not always in those words, I think. My own experience is that I built these systems and over time I realized that not all the systems actually worked for me, Right. So I'm non-binary, I use they, them pronouns. I built systems that would classify me in a binary and I literally wrote these programs. Or you know, during COVID, I was helping to automatically assign race and ethnicity to missing data fields so that we could understand whether the rollout of the vaccine was equitable. And when I put my own information into this, it came out with the fact that I was white and that I was female and all these things that I actually don't identify as right.

Kasia Chmielinski :

That could be dangerous. Funny at first, and then it's dangerous. I use a hearing aid, right? So I end up being the margin. I end up being in the gaps of most of the technology that I have built over my own career and, as such, at some point I decided no more. I'm now going to focus more of my time and energy on filling those gaps and helping others make better choices to make technology that works better for everybody.

Randy Silver:

Thank you, that was fantastic, thank you.

Lily Smith:

The product experience hosts are me, Lily Smith, host by night and chief product officer by day.

Randy Silver:

And me Randy Silver also host by night, and I spend my days working with product and leadership teams, helping their teams to do amazing work.

Lily Smith:

Luran Pratt is our producer and Luke Smith is our editor.

Randy Silver:

And our theme music is from product community legend Arnie Kittler's band Pow. Thanks to them for letting us use their track. Thank you.