Bias deep dive with Riham Satti [Podcast]

Kerry Boys: And we find that reframing really helps leaders to start to understand that it’s okay to have bias, it’s okay to talk about bias, and then we can make change.
Riham Satti: A hundred percent. And in the most basic, simplest form, it’s a survival technique, right? That we’ve been evolved.
Kerry Boys: Yes.
Riham Satti: Because otherwise we wouldn’t be able to digest all the information that we have to digest.
It’s physically impossible. There’s, there’s so many senses that we have, you know, auditory, you’ve got visual. The only way to process all of that is our brain has to make these mental shortcuts. Otherwise, it will take us a long, long time to make any decision at all.
Phil Cross: Hello and welcome to the Leaders For Good Podcast. This episode, we have the pleasure of being joined in conversation by Riham Satti. Riham is the CEO and founder of MeVitae. A UK based organisation who you will hear about very shortly at the top of the episode.
We deep dive on bias today. We talk about what bias is, what are some common misconceptions about bias? What are the implications of bias for individuals, for organisations, for society? We talk about algorithmic bias and we touch on what you and what your organisation might do to mitigate potentially harmful bias. This was a fascinating conversation. We loved talking to Riham and we think you’ll enjoy listening too. So without further ado, we bring you Riham Satti.
Welcome to the Leaders for Good podcast today. We are very excited to be joined by Riham Satti uh Riham. Hello. How are you doing today?
Riham Satti: Good, good. Thank you for having me. I’m really excited to have this discussion and debate.
Phil Cross: Likewise. So, for you, the listener, Riham is CEO and co co-founder of MeVitae.
MeVitae are a really interesting organisation and I’m sure we’re gonna hear more about that in the, in the conversation as it unfolds. But we’ll open the, open the discussion with our, our usual question, so Riham, can you tell us a little bit about who you are and how you come to be doing the work you’re doing?
Riham Satti: Yeah, of course. So my name is Riham Satti, I’m one of the co-founders of MeVitae. We were founded in, in 2014 based in the uk. We’ve got clients all over the world and really our aim is to use technology to mitigate biases. And that’s a fully loaded comment there, but I’m sure we’ll discect, and the whole story of how we arrived at MeVitae, I was pure luck, serendipity. I suppose. It was at the time, of University. In fact, myself and my co-founder of Vivek, we were both studying at, at Oxford. I was studying, neuroscience. I was doing a PhD at the time. And he was doing his masters in computer science and wanted to get a, a job at this big tech firm. I won’t name the names, I’ll just say that. And I said to him, you know, market is really competitive. How are you gonna differentiate yourself? From the world and from the competitors to get a job as a computer scientist. So we thought, let’s try and find a unique way to try and grab recruiters and hiring managers attention.
So we built this ages ago, this is probably like 2012, 2013. I’m really showing my age here. And, we uploaded this app with like Vivek’s CV on the app store, and initially got rejected. We opened up to the public, so where they can put their own details, on there. It was a bit like a LinkedIn in your pocket, right? Put your details in. You can walk around with the cv and then before we knew it had like 50,000 downloads in a couple of weeks and it was climbing a hundred every day.
Before we knew it, we were kind of like one of the top apps on the store, so we started. I didn’t know how we did it. I was still focusing on my research at the time. It was like I said, pure luck, serendipity. And then we started looking at the world of HR and just how fragmented it really is. There’s HR agencies, there’s candidates, there’s employers, and then there’s the whole concept of fairness of how do you make sure everyone gets an equal chance and realise that there’s such a fundamental challenge. And it comes down to human decision making.
So we put a task that we set upon ourselves, to try and solve that problem. How do you mitigate these biases? I did the task of getting that friend a job, he’s now the co-founder. So I did that bit. So now the next task is how do you create fairness in the workplace?
Kerry Boys: I love that. And such an interesting story and the fact you were thinking about bias before it was probably part of popular culture and language and organisations had such a huge focus on it. So I I love that you, you started out that journey very early on.
I guess given we are talking so much about bias today, it’s probably a logical place to start. Let’s unpack that a bit. When you talk about bias, how do, how do you articulate it? What do you mean when you talk about bias?
Riham Satti: That’s a really good question and it’s a, it’s a fully loaded question as well. I was looking earlier, at the definition, the Google definition of, of biases and there’s, there’s different versions and one of them I found was it’s the intonation or prejudice for or against one person or a group is fleshing in a way considered to be unfair.
That was one definition, and there was another definition that says implied or unreasoned or unfair distortion of judgment in favour or against a person or thing. But I feel like that’s a, even that’s very complicated. In, in bias, manifest in the sense that it comes from, I guess the human brain really, every, and, and this is talking from a kind, a neurological perspective.
We make hundreds of times decisions every day, and we have to process lots and lots of information. And our brains can’t take all of that, that, that information in. So we have to, I guess, bucket or group things and make decisions based on assumptions. And that’s the umbrella term of bias. Making these kind of decisions that are slightly distorted.
And there are consciousness biases and there’s, there’s unconscious biases. But they all come under the umbrella term of cognitive biases. And most people think that biases is just one thing, but in fact, there are over 150 types of biases that exist in the human brain. There’s confirmation biases, halo effect biases. There’s a, there’s a whole list that’s really, really fascinating, and that’s just the world of cognitive biases. And then you’ve got algorithmic biases that comes from, again, the human decisions that we make, but also the data that we use, the decisions machines make. And even that’s kind of another big, fully loaded term and cognitive / unconscious biases as well as algorithmic biases form the umbrella term of bias.
Phil Cross: There’s so much to dive into there and we love this topic , obviously in the workshops we run through things like inclusive leadership and it’s really a foundational concept in DEI and, I’d love to get your perspective on this because when we’re, when we talk about bias to groups of people, what often comes up are questions based on some misconceptions around the term.
Again, it’s a term that gets used out there in the popular culture, but I don’t think it’s very well understood. So, I’d love to hear from, from your perspective, what misconceptions around bias do you see? What are some common misunderstandings around the term? And, and we’ll get onto algorithmic bias in a second. Cause I think that’s a, that’s definitely another rabbit hole we want to dive in.
Riham Satti: That is a rabbit hole. There’s a lot of misconceptions around that.
Oh, where do I begin? Ok. I think there’s a few misconceptions that float around. The first one is that there is one type of bias, and I’ve alluded to the fact, there are lots and lots of them, over 150 types. And they manifest in different ways, and they change over time.
You know, one’s bias 10 years ago is not necessarily the same bias that they have now. It’s an ever evolving concept. Another misconception people have is that you can remove biases. There’s this magic wands that we have, and everything can be bias free, and we can remove biases. Misconception. Neurologically, because of the data that we process and the information we process. You can never, ever, ever remove biases. What you can try and do is mitigate those biases. We’ll try and postpone them for as long as possible. And there’s lots of methods people do in bias training. There are things that, you know, anonymized hiring in the world of HR. There are, having panel interviews when you’re recruiting lots of different techniques.
But a big misconception is that you can, you know, there’s a magic wand. You can remove biases.
Another misconception is that we all know what our biases are. I’ve had lots and lots of conversations, ‘I know what my bias is’ and I’m like, that’s conscious. That’s, you know, you’ve got those, and those are the conscious biases.
The trickier thing to solve is the unconscious ones, the ones you don’t know you have. And bringing those to the surface. And that’s very, very tricky to solve. And there’s lots of ways to try and tackle that. You know, educating, talking to people, having diverse network that you can collaborate with. Or there’s the, the implicit association tests that you can take with, with Harvard. I think that was built in the 1990s. But biases are, you know, there are conscious and there’s, there’s unconscious ones. And that’s definitely a big misconception that I, that I see. And then there’s like, there’s a whole world of misconceptions about algorithmic biases. I’m sure we said I’m gonna get into those ones.
Kerry Boys: I think one I’d just love to layer on that we see and that we find very useful when we work with organisations and and leaders is that biases aren’t always bad. So I love how you talked about bias from a sort of neuroscience perspective and the fact that it is to be human is to have to have bias and what we see with people is that bias. When we ask quite often we start a workshop and we say, what words come to mind when you think of bias? And people say judgment, harassment, racism, sexism, and all of this negativity. And, and the challenge with that, of course, is that. None of us are gonna accept we have bias when we think that bias are only these bad things.
So if we can’t accept, we have bias, we can’t mitigate it. So one of the things that we work with leaders is about understanding that neuroscience that you started to talk about and understanding that actually sometimes bias can be really helpful. So we use the really simple example of, I have a bias that snakes are bad, therefore I don’t run up and touch a snake. That’s really helpful. And actually we get leaders to evaluate their bias. And sometimes, so we talk about things like expediency bias, which is a preference towards speed in the workplace and getting things done quickly. And actually sometimes that can be really useful. If there is something that needs to be turned around very quickly, it can be really useful.
There’s obviously challenges around it with. How many people are included, et cetera. But if you’re making a conscious choice that in that moment that bias is helpful for me, then then they can be, they can be beneficial. So it’s about being aware of them so we can mitigate them when we need to. And we find that reframing really helps leaders to start to understand that it’s okay to have bias, it’s okay to talk about bias, and then we can make change.
Riham Satti: A hundred percent. And in the most basic, simplest form, it’s a survival technique, right? That we’ve been evolved.
Kerry Boys: Yes.
Riham Satti: Because otherwise we wouldn’t be able to digest all the information that we have to digest. It’s physically impossible. There’s, there’s so many senses that we have, you know, auditory, you’ve got visual. The only way to process all of that is our brain has to make these mental shortcuts. Otherwise, it will take us a long, long time to make any decision at all. And so there’s a reason for it, and that becomes really, really powerful.
The, the tricky thing is, is when you start making. Quick, quick decisions, right?
When you are, for example, screening applicants really, really quickly there, emotional responses take over, and we end up putting thing people into buckets or groups or making assumptions, and that’s when it becomes trickier. But a hundred percent, from an evolutionary perspective, we needed those bias.
For example, snakes. Snakes are bad, right? Or, you know, I’m not a fan of, I don’t know. I’m trying to think of an example. I don’t like spiders either, so I’m not gonna run towards as spider. So it definitely does help. And there’s a really good book, called Sapiens, you might have heard of it, very, very popular book, you know, bestseller book out there by, Harari and explains from an evolutionary perspective how our brains have evolved over time, that fight or flight response and how our brains have changed to be able to make efficient, rational, to some degree, decisions. So there’s definitely a, there’s always a pro and a con. So a hundred percent agree with you on that one, Kerry.
Phil Cross: And, it reminds me of, it reminds me of the, the, the sort of work that Robert Keegan did on immunity to change and the hidden competing commitments and, to sort of layer on it’s a survival mechanism.
There’s a positive intent behind all of these biases that, we developed these ways of looking at the world because they help us navigate it in a way that. Helps us feel safe or helps us make, what we think good decisions for, ourselves and who we perceive as, you know, our, our tribe, et cetera.
You know, for, for, so, you know, we might have a, we might have an explicit commitment to seek a diversity of perspectives, right? We, we go, ah, you know, I like to get a range of range of viewpoints on any particular topic, but you end up not quite doing that, right? You end up going to the same people and that’s going against your stated commitment to get a range of perspectives, the hidden competing commitment there is to keep your ego intact, because you know, you’re, being told that you’re right and you’re smart and you’ve got good ideas. And, and also to, to feel safe as well, to, to not have your, your, your sense of correctness challenged. So, I like the idea of thinking of it in these sort of hidden competing commitments, and it can be a way of, kind of surfacing these for some people.
I also wanted to point out, just join up two threads together that you mentioned as well Riham, you mentioned the implicit association test. Now, I know some people have some pushback against the IAT and, and..
Riham Satti: Oh, that’s a big debate.
Phil Cross: Exactly. But one of the one of the points people use to push back against it is, depending on say the time of day, you, you take the test, like whether you’ve eaten or not, you know, whether you’ve slept well, what kind of stimulus you’ve been, you’ve been exposed to.
So if you’re exposed to , positive views of a particular group. Cause I know the, it could, can do race, it can do gender, but if you, if you’re exposed to sort of positive stereotypes of a particular group, that that impacts your IAT. People use that as a way to dismiss and discount the IAT.
Actually what it points to is that that biases are malleable and can change over time depending on the situation. Right? So, so we’re seeing, we’re seeing like the fact that just the stimulus you were exposed to, the fact that you’ve had lunch or not and you’re hangry or tired, like impacts this, this subconscious view of people and things.
So, just wanted to thread those two.
Riham Satti: Oh, agreed. You know, it’s not by any standard like the the, the gold standard of detecting biases. But it, it creates a sense of awareness. It opens up the conversation that we all need to be having Yeah. Right now. And it’s an important topic. And you’re right, these, these biases will change, they’ll adapt, you know, and that’s something that we really need to be aware of, because we are, over time, we change, we adapt. We’re not who I was yesterday and who I am today. It’s changed. So therefore my perception, my ideas, my concepts, my opinions will evolve and therefore so will biases. And that’s, that’s why we’ve, but it does open that conversation of going okay. Where’s the starting point? Why do we have this discussion openly, frankly, and try and find ways to tackle it? Cause it’s, it’s fundamental.
Kerry Boys: I think that’s a perfect point for me to ask my next question.
Phil Cross: If you like what you’ve heard so far in the podcast and are looking for new ways to bring diversity, equity, and inclusion to life in your organisation, why not reach out for a chat. At Leaders For Good, we offer a range of solutions from diversity, equity, inclusion, strategy, sprints through to inclusive leadership workshops to DEI training for your whole organisation. So if that sounds good, drop as an email at [email protected].,
Kerry Boys: We’ve kind of talked about it, but not explicitly. Why, why is it important for organisations and individuals to consider bias? Like why does this matter?
Riham Satti: That’s a really good question. In fact, I think there’s different, ways we can tackle that, that question. There’s one of the, I guess, relating to the diversity, equity, and inclusion space. One of the main contributing factors to the lack of diversity, is comes down to biases.
Whether they’re bit a good or bad, like we debated, but it does come down to that. There’s lots and lots of research and studies out there that I won’t list all of them. That goes into the benefits of what, why you need to have a diverse and inclusive teams. There’s reports by McKenzie. There’s reports by PCG, right?
Intel that explains the importance of having a diverse and inclusive team from, you know, from a financial performance perspective, from a different perspective of thinking, from an innovation perspective. It’s not just the right thing to do, but it’s a competitive advantage. Diverse teams succeed and they do better.
And that’s a way to make sure you’ve got, so having a diverse and inclusion team makes sure that your organisation, which is on the back of having lots of people and having diverse teams strive. It makes you ahead of the game. And that’s a fundamental, thing for organisations. Organisations always want the top talent, right?
And it’s such a competitive space at the moment, and I think it will always be like that, but it’s a way to make sure that organisation are innovative, are profitable, are doing the right thing, and making sure they have the right people, the right culture in order to do that, and tons of studies to back that up.
And it’s just, we need to have these conversations time and time again. And it doesn’t just have to be at kind of the most superficial level needs to be, across the board. Everything from apprenticeships to graduates, to internships, all the way to the boardrooms. It needs to be a foundational thing, not an add on or a nice to have.
It is a structural, foundational thing, and that’s how companies can thrive and do better to be able to compete for the, the top talent. At the end of the day, companies are based on people. And people is what makes organisations grow.
Phil Cross: You touched on a few, obviously different points there and, and to kind of sum up how we talk about and think about the why behind diversity, equity, and inclusion.
But I, I think this plays into this that we could talk to bias directly with this as well, but we think about it and you, you point into this, Riham, the responsibility. The, you know, if we’re being driven, if we don’t have checks and balances and systems and processes and cultures in place, that, that, that kind of challenge our unconscious bias. We, we risk doing harm to individuals. And, and, and that’s something obviously we want to, we wanna mitigate. So, it’s the, it’s the responsibility, the opportunity you talk to a lot there in terms of, in terms of creativity, innovation, et cetera. And, and, Then the risk from an organisational perspective that the legal, the reputational, the financial risk of getting this wrong is, is pretty profound.
I just wanted to build on that from an individual perspective as well. From our perspective, it’s really a superpower to develop, you know, the, the ability to take and hold multiple perspectives to have the humility, to not think that you are right. Going into any, any situation with your opinions is the, is just the foundation of well rounded thinking.
And if leaders step into the practice of trying to look for, recognise and mitigate buyers. That actually leads to just a far more effective individual as a leader, as a contributor, as just a human being as well. So, if all those things didn’t sell you from an organisational perspective, just the fact it is a superpower that you can practice and develop is, is I think, I think worth noting.
Riham Satti: Oh, I love that. I think that’s absolutely spot on because it’s so easy for us to think about diversity, equity and inclusion from a, I guess a protected characteristic viewpoint, right? You know, I need more women in the board or, you know, I need someone from underrepresented groups. But in fact it’s beyond that. It’s that diversity of thinking of thought as well included. So completely spot on there Phil.
Kerry Boys: And I think we often get organisations coming to us, especially around bias is a really remedial thing. So it’s like we need unconscious bias training. Our teams are biased, like we have to fix this, and that is never gonna be motivating or inspiring. And I think probably all of us have been in that horrendous unconscious bias training where you walk out feeling shamed without any action to take. And we’ve really, from our perspective, this is such exciting work. You get to learn more about yourself and each other. So we really flip it on its head and take such a positive approach to how we train it and how we help leaders think about it.
Because, as Phil said, it can be transformative. It can be a superpower. Like there’s so much real, difference that you can make when you approach it from the right mindset.
Riham Satti: You can uncover new skills that you may not have known you’ve had, or hobbies that you, you’ve, it really does help from, a personal development perspective. And that is, like you said, like a superpower, right? If you can, everyone’s always trying to learn new things about themselves. So why not unlock another layer to that?
Phil Cross: And it spills over to teams as well. A, you are role modeling and B, you are opening doors. You know, starting a conversation with like, Hey, I’m gonna share my view, but this is undoubtedly driven by a bunch of my biases. And this is just my perspective. So I’m, I’m opening the door here for you to challenge, for you to chip in becuase there’s something I’m probably not seeing here. You can still share it, you can still share a perspective, you can still be a subject matter expert, but you’re doing so in a way, that again, fosters inclusivity, fosters contribution, fosters challenge, and the knock on effect is all of those good things in terms of engagement, inclusion, innovation, et cetera.
Riham Satti: Agreed. And so many, you know, organisations talk about culture, right? And making sure they have the right culture. It’s very important. And this fits perfectly into that. You know, it’s making sure that everyone has the opportunity to thrive, and be the best that they can be, but also spurring on their teammates. Right? And at the end of it, an organisation will grow and actually be, you know, ahead of the game when it comes down to, to, you know, it comes down to innovation, the profitability, those, those kind of metrics that organisations are looking for on the board level, it all comes down from a ground-up perspective.
Phil Cross: I’m gonna ask the question now, just because I’m so, I’m so keen to get onto it. The topic of algorithmic bias. So, and obviously at MeVitae, this is, this is one of your sweet spots in terms of, in what you do as an organisation, so I’m really keen to get your viewpoints here.
How should we be thinking about algorithmic bias, what is it? And, why, might we be concerned, about bias creeping into algorithms?
Riham Satti: Where do I begin? Algorithmic bias is, I think is one of the biggest guess threats that we have today.
You know, we talk about, I guess, climate change and we talk about, you know, the impact of having a diverse and inclusive team. Algorithmic biases is one of those other key threads. Because it has implications, not now, but tomorrow and the day after and the future. Because algorithmic biases are actually algorithms, that are skewed. They have biases that are due to multiple reasons, it comes down to biases that we have as, as humans that are projected, into these, these algorithms. It could be within data, right? That we’re analyzing. It could be even be the algorithms that we build or design. You know, people talk about machine learning and we don’t know what’s, it’s like a black box.
There’s an input and an output. We don’t know what’s happening in between. And, even that skews things. But then also people making decisions on the back of algorithms, right? That also sways things. And that has implications in every decision that we make, in every single possible sector in the world of hr, in healthcare, in insurance.
And if we don’t try and rectify this now. Then we’re just gonna be kind of making the situation a lot worse. And that’s why, you know, organisations are throwing millions and millions and pounds into this challenge because it’s, it’s a structural challenge from the very, very beginning. And it’s very, very convoluted.
It’s a very tricky problem to solve because you’ve got the data, you’ve got people, and you’ve got decisions on the back of that. And that’s like a, a vicious cycle that we need to somehow try and break again. You’ll never, ever be able to remove all the algorithmic biases because it’s, you know, comes down to people at the same time.
But if we can try and mitigate that, then we can make, and I’ll put kind of quotations, fairer decisions, that are important. And we’ve seen so many examples of algorithms going awry, right? A lot of, in the world of HR people talk about the Amazon situation, right? How they had their algorithms. A couple of years ago that they used their past hiring data for like the past decade or so, and machines, they used the machines to kind of determine who is best fit for a role.
And it ended up picking, a certain type of, of candidates. In this case it was, it was more men can it, so it’s, the system was kind of making the situation worse, which I think they, they’ve stopped that, that technology now. But there’s so many use cases in facial recognition. In criminal justice cases, there’s lots of situations that have arisen as a result of algorithmic biases.
And so for us, it’s one of the most crucial things that we try and mitigate it. We’re not just trying to mitigate cognitive biases, but we’re also trying to mitigate algorithmic biases for the future. So that when machines are making decisions, there is as fair as possible. And making sure that everyone is given a fair chance as possible.
But there’s lots of things that need to happen to make sure that. We’re mitigating that, before need hours and hours of discussion. But , I’ll leave that there.
Kerry Boys: Well, I’m actually gonna ask you to expand on it a little more if you don’t mind, because obviously the work you do at MeVitae is specifically within the hiring process.
So I’d love you to bring that to life in terms of the work you do as an organisation and how that links through, to the hiring process, how you see, obviously use the Amazon example there, but whether there’s any examples that can bring it to life through some of the, the work that MeVitae is doing.
Riham Satti: So for us, one of the, I guess the, our missions is, like I said, to, to mitigate that cognitive and algorithmic biases. And we made a, a very conscious decision very early on to make sure that, any technology that we build, tries to achieve that, one way or another. And that for us, we specifically focus on the, I guess, the talent acquisition process, before the interview process.
And there’s lots of, I guess, technologies out there or silo tools that you can use to try and mitigate biases. But let’s take a, a specific example, let’s say. Screening applicants. There’s lots of tools and algorithms out there to try and screen applications. They’ll might use kind of keyword matching or semantics to by be able to try and identify top talent.
And that’s all fair and well. But when you’ve got machines that are, you know, looking and reading texts and even the type of words people use, Will impact whether you get screened or not, right? Or when you’ve got an algorithms that’s taking your past hiring data and making decisions on that. What kind of people have you hired in the past and doing that over and over again times 10.
You’re gonna end up, we’re getting the same pool of applicants you’ve had before. And that’s, you know, it goes against what we’re just talking about and building a diverse and inclusive team. So that cycle needs to be broken. And for us it comes down to a series of things. We wanted to make sure our algorithms, will try to mitigate that.
And that worked in a few ways. Making sure that, when our algorithms are built, the data that we use is representative of the, the sector of the space. A lot of data is skewed, we’ll use, or it’s incomplete and that’s always gonna be the case, but it can add a lot of noise, really. So we’re very, very selective in terms of the data that we use, in our own algorithms to make decisions, just to make sure that we’re giving, an equal representation as much as possible.
Or using, a diverse, you know, we’ve got a really diverse team at MeVitae. People from different continents, from different academic backgrounds. We’ve gotta practice what we preach. We’re very, very hypocritical if we didn’t. So having our own teams, making sure that we’ve got different viewpoints.
Building algorithms. And one of the interesting things is we’ve reached out to the ICO, so the Information Commissioner’s Office, they’re also trying to use tech, you know, audit or build frameworks in mitigating our algorithm biases. And we ask them to, to audit our technology. And how are we, how is MeVitae, doing in terms of mitigating algorithm biases?
And we did that audit, late last year in fact, and it’s, you can check it out on our website, and we were ranked strongly on there’s over 10 factors that, organisations should look into to try and mitigate biases.
Transparency, so how do machines come up with the decisions that they do. It’s not just a black box. And making sure that we are transparent in saying, okay, the machine came up with this decision because it X, Y, Z. Making sure that, from a, kind of a data representation, making sure it’s as, as wide and as as possible. Making sure from, an individual’s, right, like who owns the data, whose rights are, do they belong to.
Statistical accuracy, how accurate is our machines? And we built, what we call statistical fairness tests. So we will only release algorithms if they pass these tests. And that’s part of our audits that we do regularly. So there’s lots of things that we’re doing in the background, and I think it’s really important to make sure that what we’re putting out there, mitigating those algorithm biases that I’m referring to.
Phil Cross: We might have to do a round two on this at some point because
Riham Satti: It’s a pretty loaded thing, isn’t it?
Phil Cross: There are, there are a lot of avenues we could go down, but , I really wanted to kind of highlight and, and drill into the prototype bias that we use that that can often fuel.
Algorithms with, with, in things like a hiring process. So we have this idea of, okay, these and use the Amazon example there really well, which is these are the characteristics of our top performers. So we are gonna base, our hiring decisions based on them. And, and what we get there is that conflation of correlation with causation, right?
Just because this group has these characteristics, doesn’t mean these characteristics make them successful in a role. And there’s a book I often reference that’s got a, slightly provocative title, but, but a good one. It’s that Why Do So Many Incompetent Men Become Leaders And What To Do About It by Tomas Chamorro-Premuzic, and he through all the research he’s done, points to the fact that men are often hired because they’re extroverted, because they’re kind of self-promoting in a kind of mildly narcissistic way. But because we see those as the traits of leadership, men who typically demonstrate them, are preferenced in a recruitment process. Whereas the actual traits correlated with effective leadership are higher emotional intelligence, and diligence and a greater degree of care. And that actually correlates with more effective leadership over the long term. So, so again, we have some evidence of the traits that make effective leadership, but we’re making decisions based on how we’ve chosen leaders in the past and locking that into an algorithm.
Is dangerous.
Riham Satti: It is, and it’s, a very convoluted problem, in itself, right? How do you know what good looks like in an organisation? How do you determine what your ideal candidate is? But how do you even put that down in words is very, very tricky. It’s very multilayered and, you know, you can analyze your, data, but you need to also look and understand, ok, who have been the top performers, but then actually it’s much deeper than, you know, what are those traits that you’ve seen? And it’s not, it’s not anything related to protective characteristics at all. It will be, you know, the emotional intelligence you are referring to. It could be certain skill sets that you have. And making sure. That becomes part of, you know, your requirements and your job description. So I’ve seen so many organisations write down eight to 10 bullet points of everything they need, but actually, how many of those are essential, right? Are you screening people out as a result of that? And therefore, when you’re putting it into algorithms, the people have said the right words or put the right terminology. Get filtered in. It’s easy. How much buzzwords can you put in? Right?
And it shouldn’t be a race against that. It should be a race against. How do you find potential? It, really comes down to you’re trying to find the diamonds in the rough. You’re trying to find who has potential to scale, who has potential to, you know, be the top leaders.
But we have to make sure that we’re not relying on algorithms or data that skewed. Or historical data that’s skewed because the output you’re gonna get is just as skewed, if not more. And therefore we have to make sure we break that, that cycle. Now we don’t break that cycle now it’s just gonna get worse.
And that’s why I think it’s one of the biggest threats that’s that’s out there today. We need to make sure that we are looking at the data, that we are using, the algorithms that we’re building, the outputs that are coming outta these algorithms. We need to be auditing. There needs to be frameworks in place.
And I know there’s always a lag between legal and actual technology, and that’s another debate on its own. But there should be, I guess, accountability responsibility of organisations that are also building these. To make sure that we are trying to do as much as we can to give everyone a, a fairer chance.
Otherwise it would be very, very tricky.
Phil Cross: That existential risk is really, real. Like actually as a, as a, as a civilization. Will MacAskill, who is sort of famous for his work in the effective altruism movement. His recent book, What We Owe The Future talks about different types of existential risk we face. And, and one of them he talks about is, is kind of value, values lock in. And if we only have to look at the evolution of, of kind of culture and society to know that the values we kind of aspire to or hold as the high watermark today and not the values of yesterday. And, and it’s quite hubristic and quite to think that we’ve got it nailed right now. Right?
So, and, we couple that with the fact that so much is going into these black boxes to the AI driving, you know, Google search algorithm, which dictates what information people are presented with when they, when they search for a particular query to medical systems, to systems of governance.
And, and if we’re not careful about creating checks and balances around that, again, what are we locking in? Because garbage in, garbage out, right? And, and to give that a very, kind of pointed and short example of, you garbage in, garbage out, you might remember. I think it was Microsoft, on Vale Day.
A are you gonna say the bots?
The bot? Yeah. What was, what was, what was its name? I, I forget. But they, they released a machine learning and AI bot onto the internet, and, within about five minutes it was, it was the most racist, sexist, horrible entity you could imagine, because it was just, it was exposed to that data set right.
Riham Satti: There was this really interesting, there’s a report. By the CDEI, I believe it’s the Centre for Data Ethics and Innovation. And they did a review on the topic of algorithmic biases. They interviewed lots of organisations, including ourselves and investigating what the causes of algorithmic biases are and what can we do today to try and tackle it.
And they looked at it from different sectors, including recruitment is one of those, those areas. And it does, you know, some of those things that came across is it’s the data that we’re, that we’re using, making sure that there’s frameworks in, in place. So it’s definitely something worth having a, a look at, including like, you know, even our website, we’ve got just an intro into the world of algorithmic biases, just to make sure that, cause there’s lots of misconceptions even around that.
You know, we talk about myths, algorithmic biases is a whole world of myths about that. But there is always, there’s always something that we can do to try and it’s not like, let’s shove this problem until later. There’s lots of things that we can do individually. You know, making sure that we’re not just putting more noise in, into, into data, for example.
And how do we, tackle that? So we need to make sure that we’re, we’re addressing this from a, it’s, it’s a societal thing. No one person is gonna be able to unlock this. It, it needs to be a team effort to try and make sure that we’re coming together, to try and make sure that algorithms that we use today and tomorrow are, are as kind of fair as as possible.
So I definitely recommend looking at the, I guess the review of algorithmic biases, cuz that’s a, a good way to start. Or can they look at our website as well? Of course there’s, there’s intros. But it’s, it’s one of the biggest challenges to date.
Kerry Boys: Great. I think we’ve covered a lot of ground in this, in this conversation and I’ve certainly learned a lot. Algorithm think biases is definitely not my, my specialty. So thank you so much for sharing. I think I’d love to get us to try and help, help the audience, a little bit with what they can take away. If there was one thing you wanted to, to leave them with, one thing that could do differently for themselves or their organisation. Let’s ask all of us, but what might it be if you had to just give one thing? I know that’s a really hard, hard question.
Riham Satti: I would say continue having conversations. It’s very easy for us to kind of, put it aside and think, you know, I’ll come to this later, or, but I think we need to keep having conversations with diverse people with different opinions. Different thoughts to have that sense of awareness. And that means, you know, educating ourselves and being ready to have these uncomfortable conversations. But it’s fundamental, to making sure that we are moving the dial in the right direction, and making sure that we collectively as a team talking about diversity and inclusion collectively on moving the dial.
Because to be honest, everyone deserves a chance. There’s so many diamonds in the rough. Everyone has the potential to be, you know, a great leader, a great team member, a great individual. So why can’t we all just do it together?
Kerry Boys: Yeah, I love it. I would’ve said conversations as well, so I’m gonna have to think of another one. Phil. I’ll pass on to feel while I, while I get my brain working.
Phil Cross: For me, I love this concept of, of, encouraging people to embrace the friction as a feature, not a bug when it comes to, when it comes to mitigating bias. So it requires a little more work. It requires a little more space to, to challenge the way we’ve always done things, to stop and think and, and to get that diversity of, of perspectives.
Very often the resistance we hear or see from certain organisations, particularly in things like the tech space where they, you know, there’s a, there’s a, there’s an emphasis , there’s an expediency bias at play, towards acting with speed. Is that, oh, no, we, we don’t have time for this. We, we’ll fall behind and actually, Yeah.
Seeing the, seeing the process of slowing down as, as good for people, as good for decisions as as good for the organisation. As a, as the feature it rightly is and, and not a, and not a downside. So, so that would be, that would, and, and to build that in. To, to decision making, to build that into systems and processes and actually create space, for, for that within, within teams, within processes.
That would be my, my, my ask of organisations. If I had to ask one thing.
Kerry Boys: I think my one thing now is gonna be to think beyond people. So, so often when we go into organisations, we’re asked to run training to help individuals with their bias. And yes, that is part of it. But as you’ve talked about algorithms, but also broader systems.
Enable bias or don’t. So to change something at scale, if you have a large organisation and you might need to change thousands and thousands of people, is challenging. It takes time. It’s important. But also we can make really simple changes in our process. So we’ve been talking recruitment. Let’s just mandate that we have to put all job ads through a Gender decoder tool so that we know whether we’re using. Gender bias language in our job ads. Like really simple change that we can make instantly and it, and it enables us to make change quickly and at scale. So yeah, thinking beyond individuals, it’s not all individuals responsibility. We can also look at, look at our systems and look at what change we can make there.
Riham Satti: Can I cheat and just add one more onto that .
So I was gonna say on top of that, I would also measure your data to make sure you understand the impact these interventions are. Because if we don’t track those changes, it’s very hard to know what’s worked and what’s not. And you wanna be able to know that impact.
So tracking your data across, the pipeline and being able to quantify as much as possible. We’ll make sure that you’ve got a, a very good feedback mechanism to improve as
Phil Cross: well. I’m gonna end with one more, and this is, this is the thing, this is the theme of one more, but it’s to not think of it as the one, because we’re all struggling with the one thing, right?
Because we spend our days working on this and we know it’s, we know it’s by no means one thing. There is no silver bullet for mitigating bias or, or the work of diversity, equity, inclusion in organisations. And I would encourage organisations to think in those terms too.
So if like, we’ll just do this training. Well, no, that’s a part of it. But, but we’ve gotta look at mindset. We’ve gotta look at behaviours, we’ve gotta look at culture. We’ve gotta look at systems and technology. So, to take, to step back and take a whole system’s approach to the challenge would be my takeaway.
What we, we’ll, unless Kerry’s got, got another one to add.
Riham Satti: Well, Kerry, go for it. . .
Phil Cross: Riham thank you so much. This has been such a great conversation and, and I think we can, very, very easily do a round two at some point as well. We’d just like to end as we usually do with a few getting to know you questions.
So, if you don’t mind what do you, like to explore? What are you obsessed about on evenings and weekends when you, when you’re not thinking about bias?
Riham Satti: Oh, what weekends and evenings. . I am obsessed about fairness. I really am. It’s because I’ve just seen, you know, so much in the world. I’ve seen so much kinda injustice ,unfairness. And trying to make sure that we try to be as fair as possible is something that, it’s one of my guess core values that I have in life. It’s, it’s my kind of foundational thing that keeps me going, gives me that drive, and it doesn’t feel like a, a job or a work or, or anything like that.
It just feels. This is the right thing we should be doing. And so I do spend a lot of my time thinking how do I build things that are much more fair? How do I build things? How do I give people a chance? How do I, you know, that is something I spend a lot of my, my time on. And it’s just purely because I love doing it.
It’s, and I think that there’s so much good in the world that if we can, the power we could have, if we can untap that the potential that we can have is. Unmeasurable, to be honest. So I do spend, I am a bit of workaholic as well. I’ll add that bit in.
Kerry Boys: Well, you’re passionate about what you do.
Riham Satti: It doesn’t, it doesn’t, doesn’t it feels like, you know, it is part of who I am and is, it is an extension of myself.
And, you know, if I can, if I can make the world a better place by doing that, then, then I’ve, I’ve kind of tick. My purpose.
Kerry Boys: Love it. And that’s for you. What about any other organisations that you admire for the good they’re doing in the world? Anyone come to mind?
Riham Satti: I think there’s lots of organisations doing lots of different things that are, that are great.
So, you know, Airbnb and the, the kind of advancements in the terms of belonging, for example, Nike in terms of shining a light on, you know, diverse people, for example, Microsoft and being able to, you know, use technology to empower people. Every, single organisation has something great that they’re doing.
And that sounds like a very political answer to that, but I really do believe that that every single organisation has one thing or multiple things that they’re doing that is moving that needle. And if we can get them doing more than one thing, right? You said, you know, one more thing. If we can get them all doing one more thing and one more thing, then I think we’re all, we’re all gonna succeed.
Yeah.
Kerry Boys: I love that positive aspect.
Riham Satti: I’m really naives
Kerry Boys: Mental approach to change,
Riham Satti: but I really do believe everyone, everyone can be good. And you know, if we can untap that, then the world, we can unleash potential. It’s as simple as that..
Phil Cross: I, I love that too, cuz so often it’s easy to think of the examples of organisations that are explicitly doing, you know, kind of things for social purpose, but actually if you are making a great piece of technology and doing that in an ethical way and that enables people to go ahead and do.
Do, do their life’s work. That’s a, that’s a, that’s a beautiful thing too. So I, love that. I love that frame on it. Final question. And, and this, this, this one, this one, this one’s a bit of a toughie. So , we’ll see where it goes, but what’s one thing you’ve changed your mind about recently?
Riham Satti: You know, I’m gonna, I’m gonna, you already know what the answer’s gonna be, so I’m gonna make it very, very simple.
Coffee . So let, let me expand. So I, I drink tea. Herbal teas is my go to thing. Mint teas I can have over and over again. Never used to drink coffee. We were just having a discussion around coffees and I, I always have found the taste very, very bitter for me. So strong. And I always felt like I need to add heaps and heaps of sugar into coffee.
And so I clearly, we’ve just had a discussion around coffee just before this. So clearly I need to go and try coffee again. So you’ve made a strong case, Phil, around coffee. I’ll drop an email after to let you know if my mind has definitely changed, but it’s been swayed in that direction. So I need to give coffee again.
Phil Cross: I’ll send you a link to my go-to coffee guy. And he’s a fellow brick guy, guy called James Hoffman. And he said the other day, it’s like, People tend to need sugar. Add sugar. If you’re adding sugar, it’s cuz your coffee’s, it’s cuz you probably your coffee’s bad. Like a good coffee can be sweet and not bitter and all the rest of it.
So I’m glad there’s a potential convert in the works there Riham.
Riham Satti: Yes. There’s a conversion there, so I need to, you need to send me those links and I need to invest a lot of money to some coffee. Clearly.
Phil Cross: That’s right, it’s, it’s an expensive rabbit hole if you wanna go down it. Yeah.
Riham Satti: Oh, I’ve, I’ve got into the expense of rabbit hole of tea. I literally, went to a trip to Amsterdam and they had these diffusers I brought, which were less than, I think it was like five pounds for a diffuser. And I came home and I was like, okay, diffuser it. I need tea. So I, instead of just buying, you know, one loose tea leave of like mint, I went down a rabbit hole of buying every single flavour of tea.
I’ve got drawers. Of different tea combinations now, and I, it’s become a bit, like a bit of an obsession. Talking about your first point, become an obsession. Just, you know, I’ve got minty peppermint tea, vanilla tea, chai tea, you name it. I, I’ve, when I’ve got visitors coming over to visit, I let I say the one tea, I open a drawer now and go, which one do you want.
Phil Cross: Love it. I may have come across as a bit of a coffee absolutist. But as, as Kerry knows, I, we do have, probably have about 12 types of tea at my house as well. So I, I think, I think we have more of a hot, hot, hot drinks enthusiast. So that’s one more thing collectively, I think this room has in common, we have a, a collective, collective bias towards, towards tea.
Riham Satti: love it.
Phil Cross: That’s right.
Kerry Boys: Now. I love how you linked that back to bias, Phil. That is, that is podcast extraordinaire at work.
Phil Cross: Any parting thoughts or, or asks of the audience before we close the conversation out?
Riham Satti: Hey, thank you for, for inviting. It’s been an absolute pleasure. It’s such an important discussion point and, you know, I, if we’ve changed people’s minds to have these conversations and brought more awareness. We’re on the right track.
Phil Cross: Well, thank you so much. It’s like I say, it’s been a pleasure. I’m, I’m sure we’ll find time for around two, at some point as well. We will, we will pop we’ll pop show note, links to everything we discussed in the show notes. I think we dropped a fair few book recommendations between us in this, this conversation today.
So, thank you. So thank you so much and, and thank you all for listening. This has been The Leaders for Good Podcast.
If you found this episode, The best way to support us and spread the message is by telling a friend or a colleague. You can also give us a rating or a comment on iTunes or wherever you listen to podcasts.
And if you’d like to learn more about Leaders For Good and how you can start making positive change, head on over to Leadersforgood.org and join our free community.