Insights on AI, Bias, and Business Transformation with Suzanna Chaplin

00:00 - Joanna Shilton (Host)

Today, I have the pleasure of introducing a CEO who's been in the data-driven and performance space for 15 years and is an expert in inspiring women leaders, so I'm excited to hear her insights on diversity, not just in AI, but also in sales and innovation.

00:13

But before we get into the podcast, let me tell you a little bit about her. Susanna Chaplin has founded two businesses ESB Connect and Data Labs Group, and she has been fortunate enough to work with numerous household brands, from Aston Martin to Yeo Valley Organic. Suze is passionate about leveraging data and technology to create meaningful customer experiences, and she wholeheartedly believes in the power of email as a communication channel and an ID to make acquisition as personal as retention across all channels. As the CEO and driving force behind ESB Connect, she has spearheaded numerous brand acquisition campaigns and believes in using data-driven insights to improve the results delivered for a brand. Beyond her two businesses, suze is a self-proclaimed GDPR and data geek and is dedicated to making data accessible to all, and, as part of this, she supports women in pursuing entrepreneurship and leadership roles. So, susanna Chaplin, welcome to Women with AI Hi welcome.

01:06 - Suzanna Chaplin (Guest)

Lovely to join you.

01:08 - Joanna Shilton (Host)

Oh, it's great to have you here, so we'll just jump in. Well, if people don't know who you are and what you do, maybe you could tell us a little bit about what you do, because I think you set up ESB Connect 10 years ago. So if you could talk us through that, because then you then set up the data labs group three years ago. What happened?

01:27 - Suzanna Chaplin (Guest)

So ESB, yeah, it's nearly been running for a decade now and that's suddenly when you feel like old I was saying to someone yesterday I keep being like, yeah, I went to uni last year and realized that was totally a lie.

01:42

It's been many years ago now. But yeah, esb was kind of founded around the whole idea of programmatic email and so a lot of people used email as a retention channel and email as an acquisition channel was used. But it kind of lacked any transparency in the ease to buy scale and so ESB was basically founded out of that need. So we have a large consumer base which a brand would then come to us, would create audiences on that consumer base, send the campaign for them and drive them results Very much an acquisition. I guess. Data labs kind of came down the line as we realised the sophistication in understanding the data. So the ability to kind of look at all the relationships that you can build in a data set at scale and essentially provide brands with insights on their database using underlying data sets like ESB as the foundation for it.

02:47 - Joanna Shilton (Host)

Cool. Okay, and how did you so, I guess over the last 10 years? You probably weren't using AI at the beginning, or were you Like sort of machine learning? How's it changed what you've done?

03:00 - Suzanna Chaplin (Guest)

Oh, hugely.

03:03

So I think, even like, even if I look three years ago when we started using, we started data labs and just building out basic data, speed at which we can build models, is so much quicker, just being able to scale that data in a lot more efficient ways.

03:35

I think for us in email and I always say this like AI is actually really additive to it, because there's so many reasons why an email might not click beyond just the look of the creative, and it goes back to when you send it um, whether your relationship with that inbox provider, like hotmail, gmail, etc.

03:59

Um, whether they are receiving a lot of mail into that inbox at any moment in time so they may delay your delivery, your sender's reputation, the data quality, the subject line, the time of day, what other emails are in their inbox at that moment, and all of these things are impossible for a human brain to like understand, and so you're always trying to previously make judgments of being like what's the best thing we can look at and then see if, if we tweak this, does it make a change, whereas now you can look at lots of things in one go and find out underlying patterns and remove, like, any biases that you may have of being like oh, it's definitely send time which has the biggest impact, or it's definitely the creative, etc. So that's where, like we have found, that has added a lot of value and sped things up over the last few years yeah, wow, that's amazing because you're right.

04:54

I guess at the beginning you would have been just using guesswork and we basically are you're looking for the best data point to make the best educated guess, and then you're doing a test, and even when you try and make your tests like so controlled, they're never perfect tests, so your data is never 100 accurate. That sounds weird, because what I mean by that is, even if you do a test at the same time, on the same day, you're always going to use slightly different data sets because you don't want to send the same email to the same person within the same moment, so there is always something that is different which like invalidates your results to a certain level. So it's always your best guess, whereas I definitely think ai allows you to do it.

05:40 - Joanna Shilton (Host)

Much greater scale across much more data yeah, so I guess it's speeded up the way you work and how you do. You must have so much data stored oh, yeah, a hundred percent.

05:52 - Suzanna Chaplin (Guest)

I think that's the biggest challenge in the space we're in and going back to like so much impacts, deliverability around email and then also engagement, is that we track so many different data points like when was it sent, when was it delivered, what type of inbox were they using, what device were they using, what creative, what subject line, when did they click, when did they open, did they open again, did they click? And you think about a week. We send about 50 million emails a month, so you kind of equate that out to the scale of data we have and then you suddenly are like, oh okay, that's why it's hard for like a human brain to compute that information.

06:31 - Joanna Shilton (Host)

Wow, was there any? Like I don't know, do you see sort of biases in the data? Or do you think you've kind of the human way you're trying to do it with your human brain, you've got kind of probably multiple biases. You're thinking, oh, it's probably this, it's probably that, like yeah, how do you sort of come across that? Or you know, do you, yeah, do you see any bias or not? Or how would you know? Oh, a hundred percent.

06:52 - Suzanna Chaplin (Guest)

And I think that's the biggest challenge, um one in the marketing space, um, and also like in our data itself, is because everyone has their own biases.

07:02

So if we start working with a brand, you often see it through your own lenses, like, oh, I think this would be the main USP, because you know that resonates well with me which might not resonate well with the next person.

07:16

So you're by default when you're setting up some of these tests, you're almost setting them up from a personal perspective and almost causing that bias in the data going forward. I'll give you an example. If, like someone's scheduling and they're like, well, I think monday at nine o'clock is the best time, and then they consistently always schedule around that because it gives them the best results, is that like a self-fulfilling prophecy prophecy, whereas, like it's say, on AI, you could be like split test across all the day of the week and make sure we send like every half an hour across the days, run it for four weeks and then give me the result, rather than starting from a personal perspective. And I think the reason why most people start from a personal perspective or some kind of foundation is just because we can't do as many tests from a human that you can do from a computing point that you've got to start somewhere. Yeah, wow.

08:14 - Joanna Shilton (Host)

I just yeah, because we were talking at work about what's the best time to send a tweet out or a social media post, you know, and someone was sort of going through it manually, going, well, I always send it then, and that's when and I was totally thinking that it's like, yeah, but that's because, that's because you're sending it then, so they're probably looking at it. Then why don't you test some other times and like does everyone use it? Like is everyone? Were you sort of quick to start using AI or was there any sort of like hesitation?

08:38 - Suzanna Chaplin (Guest)

Oh, I feel we were quite quick to start using it, and then there was a hesitation um I feel that was similar for a lot of people where everyone jumped into it. I was like this is really amazing. And then there was, like we discussed this on reading like that mo gold's um scary smart, yeah. And then there was that and there was a lot in the press as well that came out. Like you know, ai is going to take over the world, and so I think everyone suddenly pulled back and a lot of the big companies were also like we have to be careful of using AI etc. Because we don't know how it's going to be used. So it's almost like I feel like we did a big jump into it and then slowed it back down and now are getting back into it and seeing where we can use it in the business, which is additive yeah, because, yeah, because I heard you speak on the digital voice about that and I was like yes, I've read that book because I was going around to everyone going have you read this?

09:30 - Joanna Shilton (Host)

have you read? My god, that's scary smart. And when I started doing this podcast as well, I was like there's no one else read this. This is just really scary. And then, obviously, the more that I'm learning about AI and everything, you're kind of realizing that it is everywhere and everyone's using it and it doesn't have to be scary.

09:44

But the thing that I keep reading so I'm currently reading invisible women by um caroline cardo perez and that is freaking me out. I'm reading it and I'm just just making me so angry because it's just talking about all the invisible kind of bias or the stuff that we all just accept. You know that everything is the man is the kind of the normal and the woman is the other, and that's the bit I just sort of. I want to make sure that everyone that's using you know collecting data, that it is kind of it's new data do you know what I mean? Rather than it's sort of learning from what's happened before and all the sort of existing stuff that's out there and how biased that can be. So do you, do you worry about that at all, in sort of what you're collecting?

10:22 - Suzanna Chaplin (Guest)

oh, yes, I think like and that's what I mean from like the like, the individual's perspective, because we are essentially training it and we've seen some of the like ai, um images. You know, um, I can't remember who did that campaign around barbie um, what barbie would look like, and it ended up being like completely like one of them was holding the gun and things and people like you've got to remove this because it's so racist. But it's only racist because we're teaching it to be racist and it's learning from past behaviors and, as you're saying, there is like, will we then, without knowing it, enforce that on it? Um, and that's why, almost like ai is great at some level, but we do train it and I think, if you're thinking about building out models, you've got to think about who inputs to those models in the first place. So, and I'm a massive believer like in this and like, not just like um cultural, like um diversity, but having like diversity within your team, like from their like viewpoints and you know their backgrounds and all of those things who, um, who, uh, who did black box?

11:30

Oh, michael say you did a really good book around this. Um, I can't remember the actual um name of the book, but, um, not michael, sorry matthew. Um, but I think that's the same thing when we we're training these models, you've got to start thinking about it from different people's inputs. So you do remove that bias that exists already, because the problem with AI is you're training it on historical data sets and historical views, which will naturally have biases in.

12:00 - Joanna Shilton (Host)

Yeah, I haven't read the whole book yet, but I've got to the bit where it's just saying even urban planning is kind of all designed around men and how they would do a journey and no one's thinking about well, actually it's done by male planners and the man has got in the car and driven to work or driven to the park and ride, got the bus come back, whereas a woman is probably like when they're working they're probably taking the kids to school and then they've gone to work and then they've come back and then they've gone to the shops and then they've got to check on an elderly relative and then they've gone somewhere else.

12:28

And all this kind of stuff, all this planning for like, oh, it's two pounds a bus journey or this sort of thing, it's like, well, that's not useful, because you know the woman, if they're the ones that are providing all that care, has probably done multiple trips in the day, so it it doesn't sort of work out. Just me is kind of thinking about how you need to think how women travel as well as men travel. But that's, that's just me.

12:48 - Suzanna Chaplin (Guest)

That's the bit I've got to in the book, but I do think I always say this if I stay in a hotel room, you can tell when a hotel room is being designed by a male or female half the time, because a female will always have a mirror, like a long mirror in the room for one and good lighting around, like the dressing table. And then you go into some hotels which have no long mirrors and then you're trying to do like your makeup and it's like there's no light around the mirror and I'm like designed by a man.

13:19 - Joanna Shilton (Host)

Yeah, you can tell, every time I had a friend she was getting married in Cornwall and we she'd got the. It was supposed to be like, you know, not the bridal suite but the, the room before the wedding night or the wedding day, and um, but I knew she was getting ready there. You know, the hairdresser never was booked and there was no full-length mirror in there and we were all like hello, like where's the manager, can we speak to him? It was like, do you really think that this is the best room? Or, you know, can you get a mirror? And they end up having to like wheel some like freestanding mirror from somewhere else at the hotel.

13:48 - Suzanna Chaplin (Guest)

Uh, yeah just definitely the most frustration yeah, I mean this.

13:55 - Joanna Shilton (Host)

So yeah, I mean this is women with ai, so thinking about gender and equality and stuff, and like, do you, I don't know, do you feel that there are enough? I mean you said you've got like people in your team like quite a diverse range, but do you feel that there are enough? I mean you said you've got like people in your team like quite a diverse range, but do you think that kind of women I was happy to sort of jump on the AI bandwagon. Or do you think we need to encourage more women to be using AI?

14:17 - Suzanna Chaplin (Guest)

I think the main difference I see between the two is women feel like a fraud sometimes using AI, whereas when I speak to some of the guys in the team they're just like ashamed of using it for so many different things, whereas I feel like, women often feel like, oh well, then I'm either shortcutting or cheating. Um, and like I was saying that I muddle my words. A lot of times, grammly's come like my best friend and sometimes my emails can be long winded and I'll write a post now and then ask ChatGPT like can you either shorten this, make it punchier, or can you just check, like it makes sense? But then what do you call it? I'll then post it, but I would never get it to write the content to begin with.

15:08

And my brother-in-law was saying to me but why don't you just ask it to write your content to start with? And I was like well then, it's not my voice. He was like but it would save you loads of time. And I was like, yeah, I guess so, but you know, that would feel really like, like unauthentic to me. But then I think a lot of the men are like well, yeah, I mean, this is a time-saving, it doesn't matter, there's a purpose it needs to serve and therefore that's a purpose it's going to serve. So I wonder if there's that slight like apprehension and using it, as opposed to like anything beyond that.

15:45 - Joanna Shilton (Host)

Totally. I see that just in the people that I know that use it. And you're right. I was writing something the other day and I was like, oh, shall I just ask it or shall I try myself? Or maybe if I try, and then I get the kind of view that oh well, it's only as good as, as you say, the data we're putting into it. So if I, or the prompts that you give it, so you have to give it a really specific prompt.

16:04

But I do so, I've, I got, I got out what it had written, I was like, no, it just it doesn't sound right, it doesn't sound like me, I'll just do it myself. And then I probably spent twice as long doing it, whereas if I just used whatever it had given me, I would have saved time and I would have just, you know, would have been fine, no one would have known. But, um, I know some people think that, yeah, if you're not using it, if you're not cheating, you're not trying hard enough. And maybe that's either not trying hard enough with your prompts or, I don't know, you're not giving yourself enough time. But I do. I think you're right. I think women feel guilty about using it, but I think that's because we've only we're always having to prove ourselves yeah, definitely, and it's funny we're starting to see it in interviews more um.

16:41 - Suzanna Chaplin (Guest)

So we do like in our second interview there's normally a task and you can just we're starting to spot where people have obviously asked chat, gbt or like a similar um product to do the task for them. Because you're like the level of detail in this is just like one you just wouldn't have put that level of detail, but it does. More often it's the male who's done it and, like some are great because they read through it, they understand it. And then other people you're like you have no clue what's in this presentation. You've literally asked them, they pulled it together and they're just reeling off like the script yeah, it's good that's.

17:22 - Joanna Shilton (Host)

Yeah, that's the scary bit, I think. I mean yeah, I mean what excites you about AI? I mean how do you see it like for your business, for what you're doing?

17:35 - Suzanna Chaplin (Guest)

Oh, for our business it's just the ability to find trends and like for me, it's always been so frustrating in the email side to pinpoint really like what causes, and like deliverability issue or what causes, um, a better engagement with an email, and you can always make these like guesses, but you're like, oh, it could be this, it could be that, and so you're always like just almost scratching the surface and it's just because there's just so much data that feeds into every single send.

18:06

It's like there's so many different variables, and so for me it's like, okay, we can finally get into the depths of that and start building building up good like patterns and trends that we see that we would never be able to compute ourselves. And also in finding the right audiences for brands and we've seen this quite a few times where a, a brand will have a view on who their customer is Like. We had one FMCG brand that were very much like our customer is the rural 60 plus farmer who likes to buy local. And then basically we looked at the data completely disproved it. It was urbanites, double income, no kids yet, who were like I like to appear I buy local, but actually I just go to Waitrose or Whole Foods and buy like a major brand and we're like this might be who you think your audience is but let's just like let have like a randomized audience, basically, and see what works best and then find those interesting like patterns that you're like oh, did you ever know that you have a massive engager set.

19:18

Because the problem is like audience come self-serving, like a brand thinks their audience is someone, so they always send to those and then when they're trying to expand their market, they have narrowed it so much so they're like build a lookalike audience of this, this, whereas now you can almost be like build me a random audience and find me like people that you think might be relevant, and then, you know, test them out. And that's what like excites me. It's just kind of the ability to expand and find new trends and patterns.

19:48 - Joanna Shilton (Host)

Yeah and I guess you're. Well, are you quite transparent with that? Because I guess that's what people want. They want to know that you've got the tools and that you're using AI and machine learning and you're you know everything at your fingertips, because that's ultimately what's going to help them target the right people yeah.

20:03 - Suzanna Chaplin (Guest)

So we often like we'll say to brands like we use your primary audience, like so this is kind of like the audience you think you want to target, but let us also create like a secondary kind of more tester audience, and then let's see how they work. And then we also offer like different lookalike models as well. So it's like let's test different look-alike models based on like one being like around purchase intent rather than like demographics, and one more around like demographics. So again, you're just getting like a little bit diversity, um, in order to kind of bring in those different audiences do you think it will replace, like the number of staff that you need?

20:44 - Joanna Shilton (Host)

or I mean, just see, like the human element, because you, I don't know, I keep, I want, I want to believe that it's. You know, it's the good version. It's not the sort of you know, when mo mo's telling you like what it's going to be like and we're all just going to be at the mercy of the, the ai that's going to take over the world, that we've got to somehow retain that kind of you know, our humanness and like checking it and kind of just using it as a tool.

21:07 - Suzanna Chaplin (Guest)

So it's not I don't know, let's just freeing up our time to do something else but this is the thing one I mean it's not necessarily the most emotionally intelligent thing, so I always feel like you'll need someone to help humanize the output. So, but I do definitely think there's going to be this two-tier of like saying copywriting like um, I'm back if you ask me. Like eight years ago I'd be like I don't really see the value in a copywriter, you know. And now I'm like god, copywriting is such an art. When it's done well, it's amazing. But now you think chat gbt can output copy quite well.

21:46

But I think there would be a level where you're like well, that's obvious, it's kind of generated, and then this is like the more premium version, and so I do think it will replace some not it won't replace like you could, like people said, like it make a massive impact on like the legal in um industry because you could just get contracts written via it, and so I think there would be a certain amount of work that disappears and will be replaced by AI. But then it will encourage us to be better at a much like a more strategic, higher level, if that makes sense, and then I do always think you'll need people to help train it. So maybe it will replace a mid-section of like current work and but then does it give you more time to dedicate on other things? Yeah, definitely.

22:35 - Joanna Shilton (Host)

I did see something the other day and it was someone shared it on LinkedIn again recently and it was like I don't want AI to like be writing up my press releases and my tweets. You know, um, you know, so I've got more time to do the laundry. It's like I need someone to invent the AI that's going to be doing the laundry or the dishes, and then I've got time to be creative and paint a picture, or you know, write my novel.

22:59

So that's my worry as well is that, yeah, it's going to stop us being creative, but hopefully it will allow us to be more creative, Because we you're right we get the nuances. It's not, you know, AI at the moment, isn't, you know, sentient. It's not actually thinking this, that, it's just giving you the facts, isn't it?

23:16 - Suzanna Chaplin (Guest)

It's just spewing out data and hoping it's correct and also, like I've said this a couple of times, it does lie in things at the moment. So, like you have to really like as much as you could trust it to a certain level, I still think there's an element of like I wouldn't. I personally wouldn't feel comfortable with just being like oh yeah, I've done it on, like track gpt, I'm not going to do like any checks of it and put it out there like just automate that whole process. I still would want to go through and be like okay, okay, that makes sense. Oh, no, don't like that, change that out yeah, yeah, no, 100% agree with that.

23:54 - Joanna Shilton (Host)

Um, yeah, that's what they're kind of like. It's kind of like the benefits but the scary bits, I don't know. Is there anything that scares you about AI?

24:03 - Suzanna Chaplin (Guest)

like, please don't let that happen oh, I think it's like will they replace us? And you know everyone's like it will. And, um, this is the next phase of evolution. And I was saying like when I read that Mo Galt book I had just had like a baby and I was like what? Like I've born this baby into this world and in like 10 years time we're not going to be required, and why would they ever choose to implant in a human brain? We're like the weakest species, like they're going to put us in a gorilla. Who's stronger, like all of these things, and that massively did scare me. But then now I'm like, hopefully, I think the problem in this is how humans have to police ourselves, because inevitably we are lazy at one level, so it is easier to shortcut everything and it's our choice to use it for good or bad. But I think that's more the scary part. It's not AI itself, it's how people choose to use AI, which is more the scarier thing. Like is it going to be used to grow wealth, greed, power? I give you sorry, I'm going off on a tangent I watched this like documentary on um BBC that was.

25:23

It was all around um whether we um should allow our wi-fi infrastructure to be run by foreign companies and um, they were showing like in the future war could be. Um, um fought just using kind of like controlling our wi-fi, internet, ai, and things. So they were like saying like for example, someone could turn all the traffic lights in the country to green at the same time and like everyone would be crashing. Or if you had, like a smart cooker, they would turn the gas on in your apartment. So you didn't realize, and all of these things where you're like well, that's not the fault of the technology, it's how humans are choosing to use that for power. So I guess it's how humans are choosing to use that for power.

26:14 - Joanna Shilton (Host)

So I guess that's more the scary part of it is like how we choose to use it. Yeah, no, yeah, I totally get that. I agree completely. You're right. It's the humans that we need to be, it's how we're using it. Because I'm sure ai wouldn't have come up with that idea that, oh, I'm gonna, you know, I'm gonna do this, unless you gave it a very specific prompt. How are you going to bring you know the entire city, uk, whatever to a standstill? Yeah, that is terrifying. I feel like we need to. There's got to be. Yeah, there are.

26:41

There's lots of positives, although I was seeing they've um said this week that apple are going to be going to be putting chat gpt into the, the latest phone or the next phone. I was reading it and thinking, oh, is this going to happen to everyone? How are my mum and dad going to use this? And then I said, oh, no, it's only in the iPhone 15 or what have you. And I was thinking, but is it just going to make us less creative? But then I think it is that mix. People were afraid of the typewriter when it first came out and were worried that it would stop people being so creative by handwriting their novels and it, you know, it didn't? It changed from that into you know, the computer.

27:14 - Suzanna Chaplin (Guest)

So I'll tell you what I could see the value there and like I didn't necessarily get this when um our developer showed me the first time. But from a um, a frontend perspective on technology, ai is super useful because we're like the average user like us won't think how to speak in the technical sense. So you know, if we're asking like oh, can you find me the weather in London? Like everything's coded in binary, yes, no and things, but that's not how an average joe would ask a question. So, um, bringing ai into like interfaces allows, like an average joe to ask a normal question, um, and then the queries happen in the background, which can still be binary, but it means that they don't constantly have to change the front end interface to allow for, like, additional data points to be put in.

28:08

I don't know if I'm explaining that very well, but saying, like a traditional, like um, if you were searching for something and you'll be like I want this filter, I want this filter. So I want size a, um tops and in pink color, like that's all hard coded in, but say suddenly you, multicolcolored becomes really popular. They would have to recode the inter like interface in order to add multi-color, whereas having like ai sitting over the top means that they can make those changes in the back end without having to change the front end. Um, and understanding how like, and it will understand how those are built up over time, like, oh, someone asking like, it doesn't have to be, I want size 8 tops in multi-colour. It might be like show me your multi-coloured tops in small, which could also be size 8, kind of thing. I don't know if that's making sense yeah.

29:03

I was like that's actually really interesting because I think it will transform and actually make the internet more accessible to people yeah, definitely it's making it easier, isn't it?

29:12 - Joanna Shilton (Host)

because you're right it? The more people use it and the more we interact with it and the more people of all generations and all yeah, all sexes, all businesses, everything use it, then it is just going to get better and better and hopefully we can. Yeah, someone's regulating it somewhere?

29:26 - Suzanna Chaplin (Guest)

yeah, control it, that's brilliant.

29:29 - Joanna Shilton (Host)

I've really enjoyed speaking to you. So where do you find all your information? Like if you were going to direct anyone if they wanted to learn more about AI or more about what you do and email marketing, that kind of thing. Where do you recommend they start?

29:42 - Suzanna Chaplin (Guest)

Oh, so that's a good question. So where did I learn more about like AI? I think one just speaking to people who are using it more. And then, weirdly, actually, I found Instagram really useful. Like following people who like showed how you do different prompts and also new tools that are coming out, so they're like these are like the top 10 tools. So if you're looking for something in Excel, like these would be our recommendations or like images, and so, weirdly, I found that really useful.

30:11

And then the more I learned about like prompting, like I started to google search more like can you give me like if you wanted to prompt it for something like this, can you give me good prompt questions? And looking at like how other people had gone through scripts, and that really trained me up on that side, because I think at first I thought it wasn't very useful and then like was suddenly like oh, actually it's all about how you prompt it. Um, so, yeah, that on that side I would definitely say like. Weirdly, instagram is a very good point. Um for that. Um, on our one, I mean, I know he's available for a chat or reach out to one of our teams like more than happy to chat through on anything from like customer acquisition to also getting, like, the best out of your email data yeah, brilliant, thank you.

30:54

Where can people find, you guess linkedin linkedin um through email, um, yeah, basically, yeah, reach out to me on linkedin, I know he's there. We'll put all the links on there.

31:05 - Joanna Shilton (Host)

But um final question is there anyone that you would like to hear on the podcast or anyone that you'd like to learn from?

31:11 - Suzanna Chaplin (Guest)

oh, that is a good question. I think it'd be really interesting to see from like, a brand perspective on how so and different types of brands. So anyone at like some of the finance brands because I think ai is like almost I think it like, I think clothing retailers, beauty I can imagine using it more, whereas some of the like more traditional brands might be more scared of it it would be interesting to see the difference in like how like more traditional finance you know everything has to go through legal and data Compliance is using it versus like a more I guess like agile startup, like Gen Z company is looking at it.

32:00 - Joanna Shilton (Host)

Great Well, I'll see what I can do. We'll shout out to some brands, get them to come on. Susanna Chaplin, thank you so much for coming on. Women with AI, and thank you so much for having me with AI.

32:10 - Suzanna Chaplin (Guest)

Thank you so much for having me. It's been a pleasure. Cheers Jo.