All right. Uh Well, hello everyone. Uh I'm super excited to be here. Uh It's been a fun event so far and I've had the opportunity to look at some sessions live and what's the recording of some others? Um Without further ado, let me start sharing my screen.All right, please let me know if there's any trouble looking at my screen or anything like that. Awesome. So I'm gonna be talking about the impact of A I on the society before diving right in. I want to just quickly introduce myself. My name is Joda K sing and I work as a director of data science at Place Maker, which is a tech enabled flexible hospitality living uh platform. Uh I am also a volunteer at data science, Nigerian women and back tech and I'm linking my social media handles here in case anybody has any questions or wants to discuss any uh you know, anything on this talk or otherwise, please feel free uh to connect with me and reach out to me on Twitter if there's anything I can help on.

So I'm also working on a book right now which is titled Natural Language Processing in the real world, which is expected to release sometime in 2023. So before actually getting into the impact of A I and understanding how we use it and the very basic things uh that we interact with every single day have uh the impact on us. What is artificial intelligence? Now, it's been a buzzword for a long time now, for uh more about 10 years, approximately it's been in the limelight. Um And there are several other terms that are associated with it that you know, we hear very often. So at a high level, artificial intelligence is just the intelligence uh which is demonstrated by a computer or by a machine. So as humans, we have a particular type of intelligence and that's because we have a brain, we have this neural network connectivity inside our bodies inside our brain that emits signals. And we have a total neuron neuron structure in our brain. And you know, that sort of is part of the reason why humans have this intelligence. Now trying to give a machine a similar amount of intelligence or well not the same, but some amount of intelligence uh is essentially artificial intelligence.

Now, there are a lot of other terms which are associated with artificial intelligence like data science, machine learning, deep learning, big data predictive analytics, recommendation systems, computer vision. Now there are all these terms, it's not an exhaustive list, but all of this is uh under A I it's under the umbrella of artificial intelligence. So when you hear these terms, uh this is this is really like uh uh artificial intelligence and people talking about artificial intelligence uh to a different lens to a different bucket. OK. So we actually interact with uh a lot of the products of A I on an everyday basis in our lives. Now, we may not realize uh all the things that we are engaging with that have a machine learning or an A I powered backend behind the scenes. So for example, uh you know, when you're typing and you have these auto spelling corrections, well, that's learning from your data for sure. Uh when you're typing emails, uh if you've noticed uh very recently on Gmail, if you're typing an email, it tries to complete your sentences just based on the first word that you're putting in uh the classification of email as spam versus not spam spam.

And several other folders that especially Google has been creating like social promotions, all of that. Uh It's all powered by data. And A I furthermore, if you have noticed any social media platforms where you post images, uh and how you see uh you know, people are getting automatically tagged, maybe you've posted other pictures with your friends and tagged them manually before uh and then the next time you upload a picture, it's able to give you a suggestion or recommendation of the tag and identify people and these features are also available on like Google photos and many other platforms.

But that is also powered by A I robotic vacuum cleaners is a big one. big time A I in the back end uh little helper uh helping you vacuum and clean your house. Google translate. Anytime we see anything that's of interest to us in a different language, we just type it on Google and translate it to another language. But that is also powered by data ne I uh and then a lot of other things too. So anything you search on social media, you see these recommendations that come on the side of very similar type of content. Well, all of that is also powered by A I. And similarly, when you're shopping for something you click on a product, you have all these recommendations that come up uh that are relevant or similar to the product that you've been searching for or any other products that you have uh bought before and so on. So it looks at consumer behaviors as well. So another example which is slightly more detailed is gonna be something like this. So you know, in during the pandemic, uh online shopping increased a lot. So you may have seen something like you're clicking on a product reviews are important.

So you check out the reviews. Uh what companies and brands have started doing is classify their comments that they receive on the products uh into different categories. So this not only helps the consumer filter to the comments that are most relevant to them. So in this case, if you care about the fit of the pants the most, then you can just click on that and filter to the comments that uh are gonna you know, be talking about the fit. Uh But then it also helps the brands themselves to understand how customer sentiment uh impacts the sales of their products or what are the limitations of their products? How is the sizing like what, what are people's sentiment behind it? So that's also another uh another example of A I. So uh we've seen all these examples of what you may have seen or you do see every day in your life that you interact with, that has some sort of uh an artificial intelligence back end. Uh It's important to know how these A I models are built and this is because we are going to dive into all the great things about uh you know A I uh as it relates to our use cases in the society um and then some potential biases and improvement factors. So one common approach is looking at data that already exists and then training your machine to learn from that data so that it's able to make predictions uh for the future unseen data as it comes. So that is a very common pattern.

And what happens is the data that already exists is sometimes very old or it's something that's already happened. Uh So we call it historical data here. Basically, we information extract from the data and create an algorithm and that's your model that is making these recommendations or predictions.

There are two types of models. Uh We have supervised A I models and unsupervised A I models. In reality, there are more types as well, but this is a very popular classification of these A I models. In the supervised model, we're learning from a known input output relationship.

And in an unsupervised, we are automatically detecting patterns from unlabeled data. So one example would be uh we saw uh the image uh tag recommendations that we have seen on social media. So uh if you already have tagged a lot of your images for a particular uh one of your friends or yourself, the next time that it can recommend that this particular person in this image is so and so, so that is basing analysis on a supervised A I model because we know this particular face belongs to which person an unsupervised model would be just detecting faces uh from images or, or detecting other patterns from images.

Uh We could say that this particular person or this particular face uh might be same in all these 10 different images even though we don't know who that is in the touch of spaces uh on the different facial features that people represent. So often there are popular patterns in the data that are used to make inferences and inform model. Uh But then they're also used to uh inform model applications and we'll see what that means. There's a lot of good uh that comes from A I, there's so much advanced research that is happening uh in the medical field and in, in a lot of educational research which is tremendously helpful to mankind. Uh One popular example is image processing of uh X ray images and CT scans uh to actually detect things that a human eye might miss. So there have been great advances in research. Many brands are able to automate a lot of their processes with the help of ma learning and make sense of the large volumes of data that is being produced to help businesses grow and automate. Uh Also another factor is reduction of manual efforts. A popular example is chat bots you may have noticed on an ecommerce website. Uh When you click on the chat option, uh at first you're not talking to a human necessarily.

Uh And it's just a computer trying to understand your questions and answer whatever it's able to before actually even connecting you to uh an actual person. There are a lot of other applications in climate forecasting uh medical advancement, especially creating uh one of the recent ones.

The last few years has been creating models for speech impaired individuals uh to understand uh speech uh and any other, you know uh translation issues. Uh A lot of new technologies have been coming up as we hear about it all the time. Uh We've all already seen of some basic everyday items like your email assistants, robotic cleaners, uh home security, personalized recommendation and whatnot. So it's really touching a lot of the industry, verticals and spheres of life of our individuals now because we are using a lot of data.

There's potential for bias to get introduced in A I models. We often hear the argument that computers are impartial. Uh But unfortunately, that may not be the case, uh how upbringing experiences and culture shapes people. Uh And they internalize certain assumptions about the world around them accordingly.

Uh A I is can be looked at through the same lens. So it doesn't exist in a vacuum but is built out of algorithms and tweaked by actual human beings people. So it tends to think of the way it's been taught. Now, there are a few different types of biases uh including historical biases. So things that were true a decade ago that are not necessarily true today, societal biases, general racial and cultural biases that if present in the data that can surface to your model. And then in the end, any application using that model will be impacted as well. So A I bias bias in general is an anomaly uh in the output of a model uh that is built on top of machine learning and data. Now, this can be due to any cognitive biases. So bias is coming from the person building the model and their understanding of the the world and their data representation and the model features. And the second would be the data set itself. So if the data is not very well represented, and it only represents a certain part or subset of what actually exists in the world that can introduce another set of biases as well. So uh this is an image that I got from the British Medical Journal, but this can actually be applied to any application that you think about. So there exists a world and there's a lot of data in the world.

Now we take that data but we can't get all of the data that exists in every part of the world and uh every single user experience. So there is a subset of data like a sample of data which uh can sometimes uh make room for any discriminatory data biases of lack of representative well representative data sets. Now, once there is data, there's a lot of manipulation, feature engineering uh and a lot of steps that go into model building. So that's where uh any any biases coming from people building models or any uh data cleaning factors or data filtering factors that applied, there can also introduce the bias. Uh then once a model is ready, it's actually used for several different applications that people the world interacts with.

So when those biases make their way into those applications in those models. Uh people using it are actually exposed to that data. And as a result, it further strengthens uh the the very patterns uh that the data sample represented. Uh So sometimes there is that lack of representation and this, this is a cycle. Now, once you have such data in the world, uh anytime you update your model or take another sample, it really depends on what are the strengths in the pattern scene. And the cycle repeats, we look at some examples of A I bias. So one popular one is that if you uh on Google translate type, she's a doctor, he's a nurse and translate that to Hungarian. Uh and then translate the very same output back. It uh kind of translates that to he's a doctor, she's a nurse. Now uh for people who, you know, want to really uh change this result, there are options available to define what should be your default pronouns for any languages that have sentences that are not uh that are, that are essentially gender neutral. Uh But this is the default result.

Uh And this is because in the past, uh it's referencing a lot of past data where it was more common for men to be doctors and women to be nurses. Uh But you know, that's not necessarily true. I think these sort of biases uh flew into a modern, this is one perfect example another one would be this portrait, a IR generator app where you feed in a picture and it makes a portrait of you uh in a very authentic manner that looks very similar to how portraits were painted uh all the way back in the mid centuries.

So the the issue there was that most well-known paintings of the era were uh you know, white Europeans resulting in a database of primarily uh a particular color of people uh in the training data. And the algorithm thus was not able to take images of like Obama, Oprah Laverne Cox and make appropriate uh portraits. The the porter Air of course, acknowledges this problem. The technology behind is good, but it's just an example of a data sets that that's not well represented.

There's several other examples with actually uh detecting faces and constructing images. So one of them is open a release this new system uh that essentially turns text to image. So you input text and it will actually return an image. So it really helps you visualize what something could look like. And it's a really cool technology, but it was reported that uh eight out of eight attempts to generate images or words like a man sitting in a prison cell or a photo of an angry man returned images of men of color. Another example would be uh in a zoom background. Uh One black individual's face was actually considered part of the background and was cropping it off. OK. One popular one is Amazon's resume screening model back in 2014. So it was built uh from data from the prior 10 years.

Now, historical data contained biases uh against women since there was male dominance across the tech industry and men were forming 60% of Amazon's employees. So therefore, Amazon's recruiting system incorrectly learned that the male candidates were preferable and it was penalizing resumes that included words such as women like women's chess club champion. For example, there's several other examples across healthcare across, you know, ads.

Uh in 2021 a child ask Alexa for a challenge to do. And Alexa said, plug in a phone charger halfway into a wall and then touch a penny to the exposed prawn. So it's a pretty, pretty risky thing. Uh not talking about these biases. Uh It, it's not just to point out the issues in A I but to understand the opportunities to detecting these issues and working accordingly. So for example, historically, men have been more into sports and the participation trends today have shifted considerably, especially in the recent Olympics, there was almost an equal representation of men and women, but the advertisement and content targeting strategies are taking longer.

Uh So as a man, you may see more ads for sports and sports products. So you would uh so would your male peers, right? And now when you see on social media and what you see on social media and around you further aid in solidifying your interests uh and the most popular and the general trends for content classifications and con content consumptions further inform more and new content creation.

So thus, it strengthens the original trend and biases that leads to uh you know, longer lingering ignorance around some outlier interest that should be considered in models. And of course, bias reduction is going to not, it's gonna help every aspect that we consider here. So it's gonna help uh research technology advancements um and the products that society interacts with every single day. You know, basically speaking, understanding the context is important in these models and ensuring that you have a well represented data set. So overcoming bias uh is important and A I models can only be as good as the the person behind it and the data behind it, right? So it's only good as the brain behind it. There's a a very need, urgent need of awareness and knowledge about these models, how they are built and what are, what should be the consideration factors as a consumer of these models and also as individuals who are building these models, we need diverse A I and data science team so that there is good representation within each team such that the team collectively can build diverse uh representative models.

It's very important to understand the underlying data and think of any cases that might be missing in the data and acknowledging the problem accordingly. Uh Together we must work towards reducing bias in models. Uh Data professionals need to test models for different types of inputs.

Uh so that they're covering all the use cases. A very important concept is data drift when past data no longer applies to current situations. So models need to be monitored over time and updated and responsibility is a big factor, not just people building the models, but people consuming the models. You know, often a lot of people say that a lot of underrepresented groups uh say that they didn't dream big to begin with because successful people didn't look like them. Uh And artificial intelligence has the potential to do so much good in the world. But when it's built on top of biased data and assumptions, it can harm how people live, work um and progress through their lives. So we can uh together work towards uh reducing these biases uh by being attuned to the biases of the world we live in and challenging the assumptions that underpin the data sets we are working with and the outcomes that they offer. So we can start by reading widely engaging with progressive ideas and sharing helpful articles and research that can be used to educate others. Thank you very much. Uh It was a pleasure talking at the session. I think we do have some times.

I'm gonna see what questions are here. That is true. Uh So I'm reading Patricia's comment here. About the race corrections applied to kidney transplants. Uh I believe I did read this document as well and that's a perfect example of, you know, where these biases impact model, especially when they are uh starting to get very widely used for applications that cater to different individuals, but it's not built with considerations of different types of individuals.

Uh Absolutely, one thing I forgot to mention is that I am going to share my slide deck on Twitter. I'm gonna tweet it out uh today itself. Uh So definitely feel free to download it from there and also feel free to reach out to me if there is anything that I can help answer or anything anybody would like to discuss.