Hello, and welcome to the Retail Podcast. Now today we're going to do a bit of a deep dive in an area that has had a lot of press coverage. But translating press coverage into the reality for retailers sometimes is half the trick. I'm joined by Rob Mckendrick from REO Blue. Rob, why don't you tell us a little bit about yourself, your background, a little bit about what Oreo Blue is, and then we'll come back to the conversation around AI. Sure. Great. And thanks for having me on.
Yeah, So I'm Robin Kendrick. I work for Harry Blue. We're a consultancy in data and AI and my role is field Chief data officer. So I help our customers in terms of understanding data strategy, how they're using data and how to, you know, do things safely and in a positive way. A lot of that comes from a background of working at the Co-op for six years. We're at a variety of of data roles in ethics, data governance and engineering. So yeah, so really excited to be doing this with you.
Oh fantastic. I know Aria Blue released a report on Responsible AI. For those who've not had a chance to read it or unaware about, you know, what it means, What is responsible AI and why? Why are you guys focusing on it? Yeah, thanks. So I mean, put really simply, it is what it what it says, it's about using AI in a responsible way, which doesn't just mean from an ethical point of view. It can also mean from a business point of view, environmental
point of view. Obviously it's become really talked about as you said in the press because of generative AI and chat bots. But it also applies to things such as machine learning algorithms that you might use
for optimization or pricing. And I think that, I mean, you've only got to look recently right at dynamic pricing, which is a great algorithm for Uber, you know, where you've got an increase in demand means an increase in price, which means an increase in supply because more drivers comes that area. But obviously say, let's say concert ticketing, it's not doesn't work so well. So it looks a bit like you're using, well, it looks like a bit like being opportunistic and
greedy. Everybody's blaming is on an algorithm and saying, well, it was the algorithm that said, let's let's double or triple the price of these seats. So I think I think it's really important from that point of view of understanding how you're using some of these techniques and being mindful in the way you do that.
It's interesting, I mean, because I've seen grocers do this with users that are on applications and even at conferences, I've heard people say, you know, if you're offering everyone 10% discount, why you're offering your loyal customers who would spend the money anyway, another 10%, right? And so that was like, OK, I
don't know how it's business. And so there's this sort of line between business operations and, I don't know, profit margins and then what's the right thing to do or you know, what your customers would actually want you to do. I think that's the way to look at it. What surprised when you looked at the report?
I, I think firstly, a lot of, a lot of people are waiting for the government to say something about AI specifically and feeling like until that happens, they don't need to worry about it. But you know, CMA have announced a, a review of what happened with Ticketmaster. If you, if you use a fuse data badly, use personal data badly, you can fall, fall foul of the ICO. So, so you know you can fall foul of GDPR, right? So that that's a really easy thing to do.
As you say, if you use pricing, you can get into a situation of accidentally introducing bias where you might give discounts to one particular set of people versus another based upon something like gender or race that that might be a byproduct of your products or the way you intend to do the pricing. But you know, it's all those kind of things that people can get in trouble for, can get prosecuted for, but it doesn't rely on a new law or regulation coming out.
It's just part of the existing regulations. And I think so the first thing is people are kind of waiting for the government to say this is what you can do and before they make a change. And I think you really need to make that change now, I think. I think secondly, a lot of people think about the bad consequences of using AI as being something just in terms of brand reputation or say regulation. But a badly trained model might introduce pricing reductions which are unsustainable for your
business. It could predict, you know, you could have an AI algorithm in in supply chain, which might over purchase something because it's got some bad data or the models badly trained. So there's potential downside financial there. And also, you know, isn't it's just not cheap, right? So a large language model, which might give you a few percent uplift on a chat bot might cost you five times as much as the alternative approach. So all of those things aren't just about, you know,
regulation. They're also about the bottom line to, to grocers and retailers. And the third thing I think is that people assume that it might slow them down. And I, I always use the analogy of if you've got a, you know, racing car with a really good brakes and steering and go a lot faster because you've got the safety built in. So I, I think a lot of people worry that if they implement something around being mindful and responsible with AI that, that that might slow them down,
slow their innovation. But I think it really opens up the ability to do things and do things purposefully and, and do them quickly and well. Who would you normally be presenting this to? Who in the business should care about this? Yeah. So I think that's a really interesting thing. So I think there are a lot of people that care a little bit about it. If you talk to say head of compliance, they're they've got to worry about AI. Head of data protection would be worried about it.
Data scientists, they're kind of worried about it. So you've got quite a lot of people who have a concern but don't feel they're empowered to do something about it or don't feel it's necessarily their scope. So when we talk to people about implementing something around responsible AI, we say what you've got to do is you're going
to take this top down. You've got to talk to the executive when you've got to set some AI principles that really set the tone of of how you're going to use AI for an organisation. So, you know, go into the C-Suite, talk about, talk about some of these opportunities, potential downsides and, you know, make sure it's up to them to show the responsibility and accountability within their
organisation. So put in place some principles, put in place some cheques and measures and and then maybe enable those heads of compliance, heads of data science to do the right thing in their organisation. But you know, it really needs to come down from the top. Have you seen that in action like in do retailers do that or grocers? Are they effectively doing that? Are they trying to to do? That the the research we conducted said there were certainly very few are talking
about having done that, right. Yeah. So it, it was literally sort of, you know, one or two out of 40 retailers that we looked at in the UK have said something about that as a Co-op where I used to work. We, we worked quite a lot on that, but it was part of our whole environmental social purpose of the Co-op to do things responsibly. So we consider it partly a good thing to do. And the, the retailers we're working with at the moment, it's putting this together as a bit
of a new thing for them. So it's something that they might have been doing, but it's something that they kind of do it in pocket. So they, they haven't really got their hands around it. Last year, Global Data Science from Unilever talked about the fact they've been looking at at this worldwide and they they'd assessed 40 of their AI algorithms for being responsible huge amount.
They had 400 to go. That AI was so prevalent in their organisation that, you know, it was literally the tip of the iceberg that they'd looked at. But if I'm not a top three retailer in the UK, right, I probably won't have that depth of field or that I'm relying on other people to help me. So where do I start? Maybe is the better question. Yeah, I, I think so.
So when you start from, from that, you know, let's assume you've done the C-Suite conversation and you said, OK, we've got, we've got in place our, we're going to place our principles. We want to do something about it. I think creating a creating your understanding, your list of all the applications you might already have or systems, you know, because you might be, you might have, you know, be using Microsoft and you go, oh, switched on Co pilot. That's cool.
You know, you've got AI there. Most of the ERP systems and finance systems are claiming they've got some kind of AI capability. So you've got something there that's, you know, already in place. So understanding that to begin with and saying, what are we using? And then looking at them and from a risk perspective and say, what are the potential things that could go wrong with some of these things? What would we, what do we need to put in place to investigate
them more fully? And then you can start to work back if you've got, if you've got a system that's, you know, where we've built a, we've built a brand new system to give offers to customers and it's based on AI. Do you know if the data going into that model is correct? And what kind of ways can the data be incorrect that would that would cause that model to go badly wrong? And then you can sort of kind of trace that back.
And and as you say, not everybody has the money of a Unilever. So what things can you do? You can do things like use the algorithm, but make sure you've got some testing around it as the algorithm rolls out. You know, we used AI solutions to work out that the Co-op what, what bread to bake and what cookies to break bake. But there was a Baker who looked at that and said, well, that's, that's clearly on because I'm not going to do exactly that,
but I'm, I'm mindful. So you call that human in the loop in, you know, in the same way that, you know, in, in banking, you would hope that your application for a credit card might have somebody looking at it at some point. You know, if it's on the margin, you might go, OK, is there somebody looking at that? And in the same way you you want to see if there are any high risk algorithms, whether you've got a human in the loop who can, you can look at those things.
Is that what term? Is that a Rob term? I love that term. That's a. That's a. That's AAI term. In the, in the, in the, in the know they know, right? Yeah, yeah, yeah, exactly. So it's like, yeah, but it but it's a good one because it's like unlike a lot of the other techno jargon that that actually tells you what it is right there in that making such a. Futuristic term. I can imagine that term. It's the first time I'm hearing it, so I can imagine that term, you know, was there a human in
the loop? I was like, Oh yeah, right. We, we didn't have a human in that loop because it was just all process, which I guess then leads into I, I think what you're talking about and what the report was, is about your AI framework. So what, what are the pillars to the framework? Was that what you were just covering there? So yeah, that that was that was around the whole AI safety pillar.
So, So what we, what we've done is we've built it up from the core of data governance and architecture. So do you understand your data? Do you understand how it got into the system? Do you understand how it was collected? So that's the first, first item. You think of it like expanding out, right? So that's, that's the first, that's the first part.
The second part is looking at safety, which is where the human loop comes in and understanding how the system is built, what tools you're using and having those cheques and balances. You've got AI safety as the next thing. Building out from that, you can start to look at biosynthetics.
So, you know, just because something is doable, should we do it which can which, you know, as I mentioned at the beginning, that could be something around it's going to cost quite a lot to build this and maintain it. Is is that cost worth the worth the benefit at the end of the day, just because it's like cool, funky to have that AI, should we do it?
But also from a, you know, customer experience, as you said, if everybody gets, if everybody starts getting dynamic pricing and you know, some of the customers feel like they're being, you know, being miss sold or something that that's, that's, that's a bad thing. You can do that, but you know, it might be legal to do it as well. It's just that case of is it good from a customer safety and
ethical point of view? And then then the last pillar around that, the top thing is, is there somebody that's at the top of the organisation from a governance point of view saying, do we have a strategy? Do we have overall control of this? Have we enabled people with the right roles and responsibilities to look at this?
And it's those four areas, the data, the safety, the ethics and the governance and the components we've put within those that really says OK, and now I've got everything I need to be confident I'm being responsible with AI. Rob, thank you so much for taking us through. You know what is responsible AI, some of the outcomes good and bad, where in the business we need to to look at it.
I guess I'm curious in terms of, you know, where can people get hold of the report and what next, what should they be doing? Reports on our website, arioblue.com names on that report contains and some of the other things on the website
contain the framework. So people can look at that and understand a little bit more about the framework and, and they can make a, an assessment for themselves on how, how well they've considered some of those things in the framework around around data around safety, ethics and governance. We also have some tools that help us so we can quite quickly help people with looking at, at that maturity and, and then help them plan out what to do next. So start with the report.
There's some really great, great ideas in there about how to, how to get going. And as I say, the the framework which people can use and and have a think about themselves. I get I get this is just a personal curiosity. How long do these projects take? How long would you be? Like is this an expensive thing to be doing? I I think that that there are two things. One is the way we run that kind of maturity settlement, it's
really quick. It's a few days of interviews in somebody's office or you know, virtually more and more so, so that that's not a long process. I think that if you are an organisation who has started to think about these things and have made a mistake, then the regulator's going to be a lot more lenient on you than if you haven't started at all. So the process of getting going is really important and then you can take it at your own speed. As you know time funding allows.
You'll probably find that a lot of a lot of places in your organisation have started thinking about these things and maybe just needs writing down the things they're already doing. So you might not be in a bad situation as you as you sort of to begin with.
So having a look at that, understanding the framework and then applying it and thinking about what gaps you might have would be, you know, really first, great first step and something that will help you on the journey and help you keep you on the right side of the regulators as well. That's brilliant and definitely regulation is coming.
And so I guess it's better to be on the right side of it before it's even there exactly because it's a lot harder to unpick some of these spaghetti systems that you see. Rob, thank you so much for taking the time to to talk to us and we'll speak to you soon. Brilliant. Thanks a lot. Take care.