Nothing for Us, Without Us; Breaking Unconscious Bias in Building Products

This video features Victor Ogunjobi at DjangoCon US 2023 in Durham, North Carolina, USA.

Nothing for Us, Without Us; Breaking Unconscious Bias in Building Products
0:22:17
Published November 22, 2023
197 views

Are we really building products that serve everyone, or just a selected few?

As developers, we have responsibility to ensure that our products serve everyone, regardless of their background or identity. However, our unconscious bias often creeps into even the most well-intentioned projects and impacts the way we develop products or engage with communities, leading to unintentional exclusions.

When I first designed a Speech-Text-Analytic app, my focus was to simplify business communication with transcribed audio and valuable insights. However, feedback from the community highlighted it had limitations for diverse users, including those with disabilities who faced inaccurate captions for audio content.

This wake-up call prompted me to improve accessibility and user-friendliness for marginalized groups. And for the same reason, I'm making proactive steps to cater for the marginalized by integrating a feature that recognizes Nigerian native languages.

As tech creators, we have the power to create change, but good intentions alone are not enough. This talk will offer strategies to address unconscious bias in product and community building, while avoiding common pitfalls. Attendees will learn to understand diverse needs & experiences for more inclusive environments.

This talk was presented at: https://2023.djangocon.us/talks/nothing-for-us-without-us-breaking-unconscious-bias-in-building-products/

LINKS:
Follow Victor Ogunjobi 👇
On Twitter: https://twitter.com/chemicopy_

Follow DjangCon US 👇
https://fosstodon.org/@djangocon
https://twitter.com/djangocon

Follow DEFNA 👇
https://www.defna.org/

Video production by the presenter and DjangoCon US 2023 volunteers.

Summary

Victor Ogunjobi argues that unconscious bias is unavoidable, but its effects on products and communities can be identified and reduced through deliberate inclusion. Drawing on his work on a health-focused AI project and a speech-to-text analytics app, he explains why co-creation with medical professionals, users, and people from varied backgrounds is essential: different perspectives reveal accessibility problems, data gaps, and unintended consequences. He urges teams to educate themselves, seek feedback—especially from underrepresented groups—use evidence in decision-making, practise inclusive design, and create safe environments where diverse voices can shape products and communities.

Key takeaways

  • Everyone has unconscious biases that can affect how they perceive, design for, and work with people from different backgrounds.
  • AI and other products can produce unintended, harmful results when they are not built with relevant data and input from diverse stakeholders.
  • Co-creation with domain experts, such as medical professionals, helps make products more accurate, safe, and suited to users’ needs.
  • Teams should actively seek feedback from users and underrepresented groups rather than assuming everyone shares the same needs or experiences.
  • Inclusive products and communities can increase innovation, user satisfaction, collaboration, belonging, and the reach of a product.
  • Practical steps include educating teams, conducting diversity audits, using inclusive design, evaluating bias regularly, and giving diverse voices a role in decisions.

Summarised automatically from the transcript.

Chapters

  1. 0:00 Introduction Victor Ogunjobi introduces himself, the talk, and its focus on unconscious bias in products and communities.
  2. 1:53 Understanding Unconscious Bias The talk defines unconscious bias and explains how it can shape decisions, interactions, and product development.
  3. 3:25 Bias in AI Product Development A personal experience with a health-focused AI project illustrates the importance of accessibility, representative data, and stakeholder input.
  4. 7:19 Identifying and Challenging Bias The speaker discusses education, research, and feedback from diverse users as ways to uncover and address bias.
  5. 9:40 Benefits of Inclusive Products Inclusive design is linked to greater innovation, belonging, user satisfaction, engagement, and market reach.
  6. 11:17 Practical Steps for Inclusion The talk outlines actions such as seeking diverse perspectives, using data thoughtfully, and creating safe environments.
  7. 12:49 Workplace Diversity in Practice A sample conversation about diversity statements highlights transparency, reflection, and continued growth.
  8. 14:20 Common Pitfalls The speaker warns against assuming users share the same needs and emphasizes the value of feedback from underrepresented groups.
  9. 16:39 Inclusive Tech Communities The talk explores how inclusive communities support belonging, collaboration, learning, innovation, and equitable outcomes.
  10. 18:09 Inclusion Beyond Technology The speaker extends the case for inclusion to healthcare, government, public policy, and other sectors.
  11. 19:40 Key Takeaways The talk concludes with a recap of inclusive education, design practices, diverse decision-making, and ongoing bias evaluation.

Transcript

3,638 words · auto-generated Show

Automatically transcribed, so expect mistakes in names and technical terms.

0:21

Hi everyone, I'm Victor, I'm a digital scientist. and possibility advocates. Currently I serve as a student ambassador at School Gloria with our mission to train young brains to become curriculars. I'm sorry I'm not able to get to present in pro se I'm delivering this talk virtually. Um visa restrictions will not allow me to be there. And it's so painful not to be able to meet you, but I know someday you will be Be meeting. Um, but I'm very happy to be speaking here and to let you know how much I value the honor you've afforded me to be among the speakers for this year's conference. I decided to recall a PJ. So thanks to the program committee, program team for making this possible and seamless. You know when I was when

1:06

I was going to do the program um for this conference when I go to schedules for this conference I made it a duty to personally appreciate the program teams for the diversity and incredible work they did in creating the program. It truly shows the um support for diversity and inclusivity that the community represents. So um just a little exercise now. Um can you give them a dig into this? So um Regarding this talk, if you have any questions, you can hit me on my social media handles or my Twitter or let me say X, you can reach me out there and own Victor Gujoby. you can send me a message on LinkedIn as well. You can message me. And of course if you want to shoot me an email as well you can do that.

1:53

And I will do well to respond to as many as I possibly can So um here we go breaking uh unconscious bias and building products or let me say generally tech products or building community So I'll be using communities, I'll be using products interchangeably. Here's an outline of everything I'll be discussing in the next, I'll be covering in the next 20 -25 minutes. A spark of interest here, what actually happens when we build our products? What really happens behind the scene or after the scene when we develop our products and when we build communities generally You know, when we build products, when we build communities, when we engage in these communities, we do so not just for ourselves, but for other people too.

2:39

And the value we do that we set examples for other people. Let me tell you that again. When we build products, when we engage in communities, when we generally build communities, we do so not just for ourselves, but for other people too. And the value we do that's with certain examples for other people. And in doing that, when you call them things that work, the one important thing that we need to note is that we shouldn't do that polling Um here are some of the nugget that I put out down here on how we can understand unconscious bias in our products. One, unconscious bias refers to the automatic and unconscious associations and stereotypes that influences our that influence our understanding and actions and generally and our decision

3:25

in our decisions So if you don't mind me checking through some of my notes here, I don't want to miss some important things here. Secondly, on how we can understand our conscious bias. in our products. It's important that we recognize that everyone has unitious bias. No one is perfect. Myself I learned from personal experience on on building a product on a speech text analytic app, I learned from it and the wake-up call that I got on fixing the accessibility issues on the app was one of the important lessons I learned in my career. So it's important that we recognize that everyone they have their own unconscious bias and it can affect the way we perceive, the way we think of and we interact with other people. even people from different races, people from different cultures and backgrounds.

4:12

So unconscious bias can actually manifest in various ways. And one of the ways, or let me just list out some examples of ways where unconscious bias can play out in favoritism. You're thinking of people ahead of others in exclusions and in unequal opportunities. So I'll share you a very important story. During the COVID pandemic, during the COVID era, around 2019 to 2020, while everyone were home, myself and a few others, we We took on the project with Data Science Nigeria, Data Science Network and Data Science Nigeria DSM. We're charged with a commission to build an AI products of that good. Solve or improve local health issues in the country.

5:00

And at first I personally did not understand why they had to emphasize that. We team up with a medical practitioner the axe us the charges to include uh head tech stuff to be included in the product any project projecting rather so i was probably like why do you just make everyone do these things individually and And just get over these things, you know, as a young professional and I my thoughts was just that let me just do this thing in good and get over it. So um along the line ball as as we got practical About the project project. I realized the fact that co-creation is very important in developing AI solutions to meet the needs for everyone, to meet the needs for all users. So specifically I learned

5:45

that Um AI solutions they have unintended consequences. They have unintended results. Especially they're not designed to with input from from different stakeholders. So for example, if you have a DI powered project for diagnosis for diagnosing diseases just like the one I work on we are trying to improve or work on a new solution that will solve local your tissues. So if you have an AI -powered system that can diagnose diseases It could be biased against some certain demographic groups, and if it is not trained on a specific data set that is applicable, you may have unconscious bias playing around But you may have occasions that are screaming in. Another lesson

6:31

which is very important that I learned during the product development there and during working on the academic was that medical and health professionals they have expertise that is essential for developing these AI products. You must make sure they are accurate, we must make sure they are safe. So for example, if you have potential biases in data and ensure that the system is aligned with medical best practices. You need an L textile to guide you. It's not something you can just use your own knowledge. So I learned that. I learned the reason. I understood the reason for um data science nigeria emphasizing that we have at least a medical practitioner in the group in the group with everything to lead us on from these schools

7:19

So cooperation allows everyone, allows anything to learn from each other and develop solutions that are more likely to be successful. So cooperations help. It helps generally to make you set your paths right for developing solutions. So by working together with the medical professionals then we ensure that our AI solutions are meeting the needs and requirements for all user bases. So yes there is one next big thing here or you may have you may have asking yourself some questions and So how then can I understand or how then can I identify and challenge these biases in our products? Educate yourself, you seek knowledge in research and especially why I like conferences like this.

8:04

is that the fact that um conferences like this like Django College US they emphasize diversity in all aspects. So Make sure you seek knowledge, you ask some people, you educate yourself and team especially your team members about bias and it's and its impact. So one way is true one way through that is through conferences like this where you where talks on inclusivity and diversity generally are encouraged. The second one is you seek feedbacks from your products I'm sure as a designer, as a product designer, you don't want to miss this. You need to hear from people, you need to hear their honest opinions on what they think about your products. There is this community then I have at my disposal. There's this people I joined. One thing we do there

8:51

is when you designing your product, when you um conceiving the plan already or when you're even done, you share your product to the to the group and one thing they do there is that you receive a lot of comments you receive a lot of opinions from people okay do this do this do that do that improve on this um take care of this you're missing out on this so these are the things they they challenge you that group they tell you people from different backgrounds are are dealing with this and you know when you when you implement opinions from people from different backgrounds you can think of this bias so then i had to share my product then the speech text analytic app and i received a lot of comments and feedbacks from It which was the reason which was why I had to go back to the product and improve accessibility features

9:40

there. So, one thing again that you need to understand is why inclusivity is important the benefit of an inclusive environment or the benefit of an inclusive product generally. One is that it promotes innovation and diversity or innovation and creativity generally and it fosters a sense of belonging in the team. you know people they feel welcomed when they feel welcomed in the new community they they tend to grow they tend to contribute as much as they possibly can so again inclusivity is very important the fact that it creates a more diverse and representative user base. And second it leads to a more innovative and impact impacting products. Lastly, it fosters a sense of belonging when you welcome them nicely, you welcome people from different backgrounds nicely.

10:29

feel um they feel relaxed and they are able to contribute as much as they can. So other things that we can possibly start to benefit from inclusive products or they can make people from diverse programs. Trust me you don't want to jump with any of this is that there is increase in user satisfaction and engagement And two, you have expanded market speech and customer base. And third, we have enhanced reputation and branding pitch. We want your product to have a good impression of people. So you must make sure that it's as inclusive as possible. So generally when we say buyers, buyers, buyers, buyers, buyers, buyers, we mean excluding people. And intersectionality is another big thing. you including people interconnecting people from diverse backgrounds so when we have buyers

11:17

or more in we have intersectionality as an as an So two separate things here, bias and intersectionality. So you probably are probably asking yourself what are these steps I need to take. or that directly from conscious bias to mitigate these biases that I possibly can have or experience in my products or in community. So one thing again I mentioned is you educate yourself, you uh reach out to people, you ask questions, you make sure you're not missing out on conferences like this, because conferences like this I believe and I strongly know that is that they uh they encourage diversity from all ends because I know personally that at this conference we have people from different walks of lives here. So um different works of my brother

12:04

Do actively seek diverse perspective. You ask from people, their opinions, people from not from your zone actually, asking from their opinions. You want to hear about what people think about your products. And also as a thought that you use data default solution, making decision making rather to minimize or maximize bias, minimize rather bias and product equipment. And lastly, you create an inclusive and safe environment for people to flourish. So there could be some other ones where I believe every other one is basically under this balance that I mentioned. So lastly, we create an inclusive and safe environment for people to flourish. Okay, um I have a

12:49

some situations where we can identify and challenge these bias. Here I painted of course I'll be sharing this slide after this talk so you can check through literally So here I painted a scenario on this page of a conversation between a job applicant and let me see an inclusive inclusivity advocate. So I should say that the question here was asked by the applicants on what they would like to see in the diversity statement and the response from the manager about the diversity culture. place at the workplace was that or let me just call the conversation itself. The question is question coming from the job applicant rather is what would you like to see in diversity statements from the job applicant. So now here's the response from the advocate.

13:35

I would like to see something that honestly reflects the applicant's true experience. If someone was to write me, I've been in the environment that have not been visibly diversed, and it's relatively recent that I've become aware of how important diversity and inclusivity are to foster student success that will completely do me. So the response from the um inclusivity advocates is stating that if someone were to write this for him or her He or she will be completely blown away. I would really be impressed by someone able to be transparent about the process and eagerness to continue growing because We are all a work in progress when it comes to addressing equity and inclusion. So what the conversation is actually pointing out is that in our different

14:20

Workplace, there are some proactive steps we can put in place that would involve conducting diversity audits and implementing inclusive design practices and possibly of course fostering the culture of inclusion and diversity in our in our teams. So when we address an unconscious bias in our products or in community building together, it's important that we are aware of this common pitfalls. There's some pitfalls that are very common, which of course they're not your fault, but we need to be very careful of. Number one is you are assuming that all users they have the same needs and experiences. I mentioned earlier about my initial impression when I was working with Data Science Nigeria to work on an health-based

15:05

AI solution. My thoughts were just like Like, okay, why don't they just leave us to work on this individually? Why would they not just leave us to work on this and get over it? So one thing you should be careful of is that you avoid assuming that every user has the same uh needs and experiences please avoid that the second that is a very very dangerous thing to do please do not in not the the fact that feedback from people they are very important in steps in improving the product. If you love feedback, I'm sure, I'm even sure you're not gonna do that. So control that actively you um seek and recent feedback from people You hear from their opinions from underrepresented groups in tech or EU , probably your zone.

15:52

So now what is that big impact of encryption on products? What exactly is that big impact of of products on official rather on products. Products generally that prioritize encryption from people They have higher chances of success. So generally products that prioritize inclusions, they have higher chances of success. It's just not comparable to most videos. It's very simple as that. It's not even comparable to products that do not encourage opinions from other people. You cannot compare the success of of the products that uh that improved or that improves on um people's opinion compared to products that was just

16:39

built and that's all so we have here building inclusive the computer Why is it so important for us to be take communities that are very inclusive? Just like you travel to a place that you have not been to before, you travel to a country. I have never been to before. It feels so strange for you to even interact. Honestly, for most people I I would say it's very difficult for you to interact with people. You can't just meet people the way and talk to them. them so when you travel to a place you've not been to before you feel so strange you feel like a stranger which is not your fault so so also in the case of inclusive inclusive tech of communities. It's important. Inclusive tech communities generally is important because it creates a positive and supportive environment for people to flourish where everyone feels safe, where everyone feels welcome and valued. It

17:24

generally fosters diversity. promotes innovation and lead to better products and solutions. And here we have some benefits that it provides opportunities for collaboration, knowledge sharing and learning from diverse perspective. it's attracting wider than it's food and improve employee satisfaction and attention and generally it contribute to a more equitable industry. And very importantly we have a lot of good leadership in these communities. It's very important that we have good leaders because they play a very crucial and important role they setting the tone. What do I mean by setting the tone? I mean the culture and the values of the community. When you have good leaders, you set good tool for these things in these communities,

18:09

in this environment. So um one big thing here again is inclusion beyond the tech industry. You know, at conferences like this that we say that increase inclusivity is um apparent and you know since since the beginning of this talk I've been saying tech this tech tech tech tech take and it seems as though we have inclusivity or we promote inclusivity only in tech industry No, even in different profession in different sectors in the country, we have in governments and healthcare and public policy Inclusivity also is something that should be encouraged. So um when I was mentioning earlier about an experience of working on health-related projects, that is basically tech in health

18:55

and the same thing health employee tech skills so they work together so it's not only in tech that we have impossibility to be encouraged we need to have inclusive inclusivity beyond the tech industry beyond the borders of tech. So if you're here and you're not a tech person, you're not tech is not your zone, rest assured that you need to uh include You need to promote diversity in your teams as well. It's not only tech that we need to have that encouraged. So let me say like a recap by summary of everything that I've been saying for the past 10 to 15 minutes now. I mentioned at the beginning that addressing unconscious bias, addressing these biases is very crucial for building

19:40

inclusive. and communities. Secondly, I mentioned that understanding diverse needs and experiences is essential for creating more inclusive communities and more inclusive products. Third, I mentioned some common pitfalls that can we can look, we need to look to help us in mitigating these biases in product development and community media. And I mentioned some proactive steps that we can take to mitigate these biases. Number one, again, consciously educate yourself. Consciously educate your team members on why it's important that we have inclusive products, why it's important for us to encourage inclusivity in communities and one good way that you know of course

20:27

is when you attend conferences like this that diversity from all perspectives is encouraged you feel a sense of um entier me feeling sense of okay I must do this I must encourage this in my team and also you need to implement inclusive design practices to ensure that your products meet the needs of every person And you create an opportunity for diverse voices to be heard and included in your decision-making processes in your workplace. And also regularly Educate, evaluates and address bias in product development and community building efforts. So at the end of this slide, I attached a demo video of my speech text analytics and just like it. One minute or two

21:13

minutes video. I attached demo video so you can do well to check it out. Not a full-fledged project actually, but can do well to check the demo out. So I think I should Put again today's now once again you can connect with me on my social media handles you can check out my LinkedIn Victor Ujbi and Twitter X, as we have it. X, you can connect with me and reach out to me personally if you want to ask some questions. We have some things you need to discuss with me generally, you can reach out to me. And I'll do well too. um reply to as many and as many as possibly can. You can also post along me, you can also tweet the Baway and my talk using the general hashtag. So once again I would like to take the program

21:58

community for making this possible and for making this registress. So sorry I'm not going to do that in person, but I believe you are going to also in time with me. So thank you all for having me engaged in this conference.

Questions this talk answers

What is unconscious bias, and how can it affect the products I build?

Unconscious bias consists of automatic associations and stereotypes that influence how people perceive, think, decide, and interact. In products, it can appear as favoritism, exclusion, or unequal opportunities for people from different races, cultures, or backgrounds.

Discussed at 2:39

How can co-creation help prevent bias in AI products?

Co-creation brings together different stakeholders, including domain experts such as medical professionals, so the product reflects the needs of all users. This is especially important because AI systems can produce unintended consequences or become biased against demographic groups when they are trained without suitable data and perspectives.

Discussed at 5:45

How can I identify and challenge unconscious bias in a product?

Educate yourself and your team, seek honest feedback from people with different backgrounds, and use that feedback to improve the product. The speaker’s speech-to-text app, for example, gained accessibility improvements after he shared it with a group that challenged him on what was missing.

Discussed at 8:04

What are the benefits of building inclusive products?

Inclusive products encourage innovation and creativity, create a sense of belonging, and represent a more diverse user base. They can also increase user satisfaction and engagement, expand the customer base, and improve an organization’s reputation.

Discussed at 9:40

What practical steps can teams take to mitigate bias in products and communities?

Teams should actively seek diverse perspectives, use data to inform decisions and reduce bias, and create an inclusive, safe environment where people can flourish. They should also educate themselves and regularly evaluate and address bias in product development and community-building work.

Discussed at 12:04

What common mistakes should I avoid when designing for diverse users?

Do not assume that all users have the same needs or experiences, and do not dismiss feedback. Actively seek and respond to feedback, especially from underrepresented groups, because it can reveal problems that the original team missed.

Discussed at 14:20

Why are inclusive tech communities important?

Inclusive communities give people a supportive environment where they feel safe, welcome, and valued. They foster diversity and innovation, enable collaboration and knowledge sharing, attract a wider talent pool, and contribute to a more equitable industry.

Discussed at 16:39

Note: We understand that names change, people change, and bodies change. We respect each individual's journey and privacy. If you have any concerns about a video or need us to remove content, please don't hesitate to contact us. We will handle your request with care and promptly address any issues.

More videos from DjangoCon US