XClose

UCL Minds

Home
Menu

WWWAI - Episode 3

In this episode, host Professor Rose Luckin (UCL IOE) discusses how UK businesses - small and large, and across different regions - can engage with AI and attract tech talent, with guests Vimla Appadoo, Head of Experience at Culture Shift, and Dana Elman-Vishkin, Innovation Delivery Manager at Digital Catapult.

Episode 3 - Transcript


Rose Luckin
Hello, and welcome to the UCL and British Academy podcast series working well with AI. I'm Rose Luckin, Professor of learner centered design at the UCL knowledge lab. In this podcast series, we're exploring how artificial intelligence AI is changing the world of work. AI has long been predicted to reshape our working lives, and it has developed in leaps and bounds over the past decade. And as we emerge from a global pandemic, we're rethinking how we work, what sort of work we value, and what we need for the future.
[music fades]
In this episode, we'll be discussing sector and scale. We'll hear about the experiences of UK businesses as they engage with AI, especially those in less oriented sectors, and small and medium sized enterprises. We'll explore how the impact of AI is felt across the different geographies in the UK and finally, we'll discuss how UK businesses can use AI to tap into more diverse talent pools and create more diverse workplaces. 
And today, I'm so pleased that I have with me Vimla Appadoo, head of experience at Culture Shift, also a board member at the Greater Manchester Local Enterprise Partnership, as well as a trustee for diversity UK. Welcome Vimla. And Dana Elman Vishkin, innovation delivery manager at Digital catapult and former senior policy advisor for innovation, enterprise and trade at the Federation of Small Businesses in the UK. Donna has also worked with corporate clients and an analytics consulting firm that has since been acquired by Deloitte analytics Israel. Interesting! And for both of you, my first question, and I'll come to you first Vimla is, can you tell me what really gets you fired up, gets you out of bed in the morning to get you working and thinking about AI?

Vimla Appadoo
Yeah, absolutely. I think, well, first of all, thank you for having me on the podcast. It's great to be surrounded by such inspirational people! So the thing that really gets me fired up around AI is what you mentioned in the introduction, and that's the diversity piece. It's how do we leverage what artificial intelligence can bring to the workplace and use it to help overcome some of the inclusion barriers that we have? So historically, technology has excluded disenfranchised communities, it's been very exclusive in who's, who it’s included and how decisions have been made. And I think we're on the brink of AI becoming more commonplace in the workplace. And therefore, there's a huge opportunity for us to overcome some of those barriers. And really, as practitioners in the space use that opportunity to make sure more people and more voices are included.

Rose Luckin
Thank Vimla, and Dana, the same… What sparks your interest, gets you going, gets you fired up when it comes to AI and the workplace?

Dana Elman Vishkin
So for me thinking about AI starts with the importance and use of data for small businesses. Because of my background… Before specializing in small businesses, I worked in analytics in Israel, where I'm originally from, and you may remember the phrase a few years ago, the data is the new oil. I think there's something a bit… It's, it's a bit misleading. The exciting thing about AI innovation is collecting the right types of data in the best ways possible. And we want to ensure accuracy when we use AI. It's a brilliant tool, which also provides, um, generates social impact. And I really became excited about the technology when I saw how social impact is generated by AI startups in various industries, including healthcare, agriculture, and it's a pleasure to be here today, the podcast, of course today to speak about that. 

Rose Luckin
Thank you. And it's a pleasure to welcome both of you to this podcast and I think I can understand both of those reasons for being fired up about AI!
A House of Lords inquiry in 2018 highlighted the importance for the UK business sector as a whole of being able to compete with other countries in the development and adoption of artificial intelligence in the 21st century. So it'd be great during the first part of the conversation today to focus on how UK businesses from different sectors and sizes are doing on this process of AI adoption and development. Particularly small and medium sized enterprises and those businesses who have not traditionally engaged with AI in the past or perhaps not even with digital technologies. 

So Dana, coming to you first, if I may. And I'd like to ask you about your current role with the Digital Catapult, because you interact with a lot of UK business working to develop AI. And you've also worked extensively with small and medium sized enterprises adopting AI in the past. So, I'm intrigued about your time at the Federation of Small Businesses, when you authored a report in 2018 that touches on small businesses’ productivity, digital technology, and AI. So what do we know about UK business AI adoption and development in the UK? 

Dana Elman Vishkin
That report was a really interesting piece that I am really pleased that I had the opportunity to conduct. Just quickly about the reports: it was conducted in 2018, as you said, and it was conducted through a survey that was sent to all FSB members across the UK, and members were invited to participate through email and social media channels to participate in the survey. We also conducted semi-structured interviews and focus groups across the UK in Cardiff and Glasgow as well as in London. We wanted to test what is innovation for small businesses. And when we tested innovation, how they engage with it, what it means to them, we realized that for some businesses, it's the ability to compete on a day-to-day basis. And for others, it's a buzzword. 
So the first major findings from that report was that of all FSB members only 2% have used AI in the three years prior to the survey. And most of the firms that have said that they used AI were a bit, were in the larger, in the larger categories, so above 10 employees. We also saw that of all members in the information and communications sector, 15% have adopted AI. So we see that from a sector when you analyse it from sector point of view, there are sectors that use AI more than others. We also found that 4% of members in manufacturing adopt AI and 2% of professional scientific and technical activities industry adopt AI, whereas none in construction or retail have done so. 
I think it would be useful to provide more context about digital technologies’ adoption in general because AI is an advanced next step after adopting digital technologies. I'm happy to add another finding and that's about views on AI among business owners. So, we found that 32% of small businesses think that AI will provide value to their business and this positive view was, this positive view has increased with age of business owner. So, if the business owner was up to the age of 44 years old, we found 47% of them had a positive view on AI and their impact on the business. And the conclusion from these findings were that we mentioned in the report was that the UK should prioritize specific approaches to make sure that all groups can have the ability to adopt AI. But it's worth knowing that one policy to fit all is not necessarily the right approach, because these are distinct groups.

Rose Luckin
I can see that from the data that you've described, that is fascinating. I often wonder to what extent, in a part of the world where I look most of the time, which is education and training, it seems to me that so many organisations are not aware of the possibilities that AI could bring to their businesses. And yet others are, so there it's very varied, I can see that. And that's really interesting, the data from that report. Thank you. 
So Vimla, I'd like to come to you now, because you've done a lot of work with the UK, digital and technology sector, What have you witnessed through your experience within your own business, and through the work you've done with SMEs? Especially those who are adopting these technologies for the first time.

Vimla Appadoo
Thank you, I'd really like to pick up on a point that Dana made there around AI being a bit of a buzzword.

Rose
Please! 
Vimla Appadoo
And I think it's a really, really important point, because it shows us when we're working with SMEs or in the technology sector is the lack of understanding about what artificial intelligence actually is. And often what we start to see, what I've seen in my experience is that many businesses, particularly SMEs, or startups, will say that they're an AI company. But in reality, they're just working with data. And there is this tension between organisations that understand data and using data versus using artificial intelligence. And that you said, like you've explained it, that next step, and that next progression in digital transformation… I think that misunderstanding or misconception of what AI actually is, is one of the biggest barriers, and that holds the key to unlocking what the future is for artificial intelligence until we are able to get that clarity and understanding what it is and the benefits it can have. We're not going to be able to kind of nurture that sector in that environment.
And as part of that, I think the, again, going back to that diversity and inclusion, it's understanding how we collect that data in an ethical way to ensure that all voices are included in that process so that when we do get to that AI part, it's, it's therefore representative of the communities that we serve. And at the moment, I think, again, one of the barriers is we've got two, two kind of trains going at the same time. We've got the slow adopters who don't understand AI, and therefore, they're just starting to unpick the data that they have and how to use it and how to do that kind of simple data analysis to make conclusions, versus the fast train of data collection and AI in the big corporates who are pulling data from all these sources aren't questioning the validity, the, you know, the ethics, the who's been excluded, what does this represent? How do we show what the conclusions are? And are just running with it to prove that they got in that space, they are the leaders, as you said, they can be the kind of forefront for the next 100 years. 
And that's a really big tension, because it sets a precedent, it sets the kind of traction in the direction of travel for those smaller businesses who are trying to play catch up. And if we don't slow down the people that are moving fast and speed up the people that are moving slow, we're going to end up in a space where AI… I don't want to say is dangerous, but just isn't doing what it's meant to do. And it isn't having that social impact that Dana spoke so eloquently about. It can really quickly have the adverse effects. And we've seen that through AI technology that has been launched in the world, over the past few years of just not quite being fit for purpose for masses of, like masses of the population. And that's one of the real dangers. 
But I also think that one of the barriers for the digital technology sector is access to talent as well. So in Manchester, we're really lucky to have such a huge talent pool, with kind of the three universities there. And that leads to graduates from data science backgrounds, or computer science backgrounds, and often PhD students, you can feed straight into that sector. But that isn't true for regions across the UK. So we need to understand how we work together to help retrain people, upskill, or whatever it might be to ensure that talent is there for what is quickly becoming the future of technology in the space.

Rose Luckin
That makes a lot of sense. I can understand what you're saying about talent, and only too well. But coming back to your point about geography, and particularly how we might encourage AI development and adoption outside the “golden triangle” of London, Oxford and Cambridge. We know that the UK Government spends less on research and development than the OECD average, and that almost half of the existing spend is spent in that “golden triangle”. So clearly, there are some, some problems in terms of geography and different regions. And so thinking about, you know, the impact of geographical location, in terms of the differences in AI innovation and adoption across the country. You've obviously worked across the country, but what have you seen in particular, with your work with small and medium sized enterprises outside the “golden triangle”? You said a little about Manchester, but can you say a little bit more about that, please? 

Vimla Appadoo
Yeah, so I definitely think that within, within Manchester, well across Greater Manchester we've been able to foster a community in and of itself, and initiatives like the Northern Powerhouse are focusing on what it means to be a technology business outside of London has really helped that kind of community thrive. I also think it's one of the knock-on effects of COVID has shown how easy it is to promote work and, and to build those connections and relationships with London or the “golden triangle in that virtual space. So, without needing to be physically based there, which, again, has really opened up the doors for what that interconnection looks like, but also open the doors for hiring outside of your region, because you can work remotely now. But yeah, I think it's really, it's really important that we keep that momentum going and don't fall back into staying within that “golden triangle” and staying within those barriers but continuing to use online tools as a way of connecting and establishing how AI innovation can happen outside of it, but also leveraging everything that is happening outside of London to make sure that that communication is fed back.

Rose Luckin
That makes sense. And it's interesting, the point you make about actually the pandemic and remote work perhaps is helping smaller companies, you know, have greater opportunities because they're not perhaps having to pay for office space. And they can access talent, from lots of different places. So that's really interesting. It would be nice to see that kind of benefit coming out of the pandemic, for sure. And Dana to come back to you please. In your report, you touched on the importance of the UK government's coordination and stewardship role to support a wider diffusion of innovation and AI uptake across the UK. So, what type of policy recommendations do you think would be useful for the UK Government that could encourage firms to modernise and work with digital technologies and AI? Especially for those companies outside the “golden triangle”?

Dana Elman Vishkin
Yeah, so if we focus on the adoption part, I would say that we should, when we think about small businesses as AI users, the end users will buy an off-the-shelf product, or will, will assess their needs and implement bespoke tool. The more traditional small businesses, we're looking at: nurseries and makeup artists, small accounting firms, we're looking at those more traditional companies. And there are several options that can be considered by the UK. So, first of all, the “Help to grow” scheme. It’s relatively new, it's a good step forward. And the help to grow scheme is… Further work can be done by increasing the grant value to achieve adoption. So, for example, the “Help to grow” is a grant (I'm talking about the digital track, digital adoption). A small business founder engages with an online platform to adopt digital technology. But I think it won't replace the need for a track with an advisor. So still, there's a need for that. 
There's also the thing about innovation vouchers or creative credits, it depends about the specific branding, but we're talking about a form of direct matched funding, which is a good incentive. And I saw that there is an existing scheme in Birmingham and other cities, but there is a list of cities. So, if a small business does not come from that list of cities, they don't, they're not eligible. And I know that there is another scheme in Manchester, but it will be great to see wider outreach for this scheme because after the pandemic as Vimla said, it's important to see how businesses can come from all across the UK and, and connect remotely with, whether it's with an advisor or an online platform, and see how they can use this great tool of AI to promote their needs. 
And I think that there's another element that I'd like to touch upon and that's the signposting. How do small businesses know about what exists out there? So, from the reports at the time, we found that only 10% of innovators accessed government support to make significant, to make changes in their business. And of those that did not, almost half said they did not know of any support. So I think this is an interesting finding. So back in 2018, there was definitely further work to be done. I think the UK is doing much better today. And, and I'd like to… I hope to see improvement in signposting after implementing this new innovation strategy. Now that adoption, diffusion of innovation is ingrained in the UKs objectives.

Rose Luckin
And now I'd like to move on to the topic that I know is very important to both of you. And that is diversity and equity in the workplace, and particularly in UK businesses. And this is central to any discussion of AI in the workplace, and indeed, any discussion of the wider technology sector. But I'd like to get a sense of how it plays on the world of AI and work and how might the adoption of AI help UK businesses address the issues that they face with respect to diversity and equity. We have touched on this a little bit already. But Vimla, you've done a lot of very important work on diversity and equity in the workplace, from being trustee at Diversity UK, delivering fearless change through culture and experience and have spoken numerous times about the need for a shift in the diversity we see in technology. So how can AI help UK businesses tap into a more diverse talent pool? 

Vimla Appadoo
Yeah, I think there's a really interesting, it's a really interesting tension in the recruitment space with how AI can be used. It’s very chicken and egg in that the way we've built a lot of HR tools so far, that are based on AI, have inbuilt bias into the systems in that space. And that's because we have a lack of diversity in the technologists and the data sets that are feeding the algorithms. And so therefore, it perpetuates the problem. However, we're not able to overcome that problem until we get more diversity into the sector. So it's this cycle of kind of, what do we do? And I actually think that's the role of practitioners to step in and say we're not ready for the AI technology yet. Because actually, as a sector, we're not mature enough to understand how and what we're building is impacting or disenfranchising those who are already disenfranchised. And that's a big pill to swallow in the sense of, we've got it, we've got the technology here, why aren't we using it? And it’s like actually it's just not fit for purpose. And often, we don't realize it's not fit for purpose until it's too late. And we've rolled it out and understood the implications. 
And that goes back to the human centered design approach that we need to take of testing it early, understanding where those problems come. But including users in that process, the end users but also the practitioners very, very, very early on in that process. Whereas at the moment, it happens far too late. And it does mean a really big shift in our mentality and methodologies of how we're developing technology and putting it out into the world. So at the moment, I'd actually I, I'd push back and say, I don't think AI can help us get into a diverse tech pool just yet. I think we need to relook at our policies and processes, and go on a culture change journey ourselves, as people recruiting into technology to understand where those barriers exist in the first place, before we start looking to technology as the solution for overcoming those problems and biases. Because equally, there's no point hiring diverse technologists if they walk through the door, and then, and then face an unsafe working environment, or microaggressions, or aren't able to operate through the business or aren't given the same opportunities, as their colleagues. So, until we actually address the systemic entrenched behaviors and patterns that we see that we know are barriers to minority groups, or anyone who doesn't fit the norm of the technology sector at the moment, until we address those, technology won't be able to be a solution.

Rose Luckin
Yes, and it's a worry, isn't it? If I understand what you're saying correctly, you know, we can't just tweak the system by adding some AI at the recruitment, and we have to do something much more holistic in terms of looking at diversity and equity within the kind of work culture, as well as at the stage of recruitment. And that's a big piece of work. I'd be intrigued to know whether there are ways in which AI could be brought in to help with that, I don't know. I mean, I wondered if some of the work that you've done gives you any kind of hints as to where AI might better be used rather than the recruitment end.

Vimla Appadoo
Yeah, I've done a couple of projects around greater Manchester, I'm particularly around the black and minority ethnic experience in the technology sector. And our ambition with that project is to collect data at a GM level to understand those barriers. But again, it's not deep tech, and it's not big datasets. It's more the simple technology and understanding where those commonalities lie, and then using it to inform better policies and processes. And as someone who isn't a technologist, I don't know how we can translate that into an algorithm or into something that does that for us. But I think there is an opportunity to do that, if you were to then scale that out across different regions and understand, like, take it out, lift it out almost but I definitely think technology does play a role in it. It's just not quite there yet. And we need to understand how we leverage the existing technology we have around blind recruitment or you know, there’s, there are really interesting tools that pick up misogynist language in job specs and recruitment efforts that can help us level the playing field around gender and all of that kind of thing that can help us on that journey. But as you say, the hard work is actually the culture change, it would be so much easier to just say “this AI does it for us”.

Rose Luckin
Yes!

Vimla Appadoo
It's just not that simple, until we're able to, to change our own mindsets around it first.

Rose Luckin
And that sounds like a case for human, not just human centered design, but also for interdisciplinary cross stakeholder groups getting together. Because you were just saying you're not a technologist, but you understand a lot about the nature of workplace culture and some of the issues that the technologists wouldn't understand. So, it's also an argument for that kind of cross stakeholder approach, I think as well, isn't it? 

Vimla Appadoo
Absolutely. 


Vimla Appadoo
But I also think we need to go further upstream and hold, kind of PhD panels that are accountable to the datasets that they're asking their students to collect. So rather, rather than focusing on the outcome and conclusions, it's actually what was the data that went into this? And have you included a diverse range of people? Have you been able to show that you've gone above and beyond to, you know, work with communities that are often left voiceless in this process? And until we start doing that early or upstream and encouraging it as part of that academic process, I think we will continue to see these problems later downstream when it gets into society, and it gets into the end users’ hands and you know, they feel the effects of that. And again, for me, that's just a consciousness, it's, it's adding in that step earlier on to make sure that we address it before it's too late. And like you say, before AI continues to get a bad rep of not doing, not being fit for purpose and not doing what it can because there is such a huge opportunity and potential for it to make a massive positive difference.

Rose Luckin
So, it's not just important, it's urgent as well, isn't it? amd definitely needs attention. Thank you. And Dana, just I'd like to come to you and ask for your views around diversity and equity with relation to AI. What's your perspective?

Dana Elman Vishkin
I think that responsible AI is not the same as AI. And the conversation is gradually developing and becoming more sophisticated in the UK to, to mitigate some of the negative aspects of AI. Because, as I said in the beginning, we're not just enthusiastic about the AI as technology, we're looking at the actual accuracy of AI, we're looking at making sure that it will add value. And I see this as in evolutionary process for technology, for such a technology. 
So, for example, at Digital Catapult, we engage with dedicated experts in product management and human centered design for AI. We expand our AI ethics capacity through collaborations. We continue our investment in tools to aid the development of responsible AI. And we, we offer and consider scalable engagement with the public through webinars, because we want to make sure that we're going, we're going in the right direction with responsible AI. So I, I think that time, is… With time we see progress in that. I agree with Vimla that there is definitely further work that needs to be done. I'm optimistic, I'm optimistic that we'll get there.

Rose Luckin
I like optimism. And I love the fact that you're also trying to help the public understand, because I think that's part of that, that's sometimes, you know, missed out and not giving enough attention. Well, we've certainly had a really interesting conversation, and identified some important current issues, and touched upon some of the trends that could have a major impact on the future of work in the UK. But I think we can't end the podcast without, and we've had a positive note already. That was great to have that optimism Dana! But I'd love to ask you a little bit about the future. 

Dana Elman Vishkin
The world I'd like to see in the future is a world where we have more diverse and inclusive groups of small business founders. That we see more women joining this entrepreneurial spirit. And I am already… I'm optimistic not just for the sake of optimism! I read a report, a recent report by Coursera, about upskilling and many women, much more than before, many more than before are studying and upskilling in AI. So I hope that yeah, I hope that a world where you keep businesses make the best of AI looks like a world where businesses embrace AI as the embrace the worldwide web. Where it's easy to understand it's accessible. And it provides value to the business.

Rose Luckin
I couldn't agree more. I think that would be great! Thank you, Dana. And Vimla, for you, what would it look like if we had UK businesses making the best of AI? How would that look?

Vimla Appadoo
I think it would feel happier!

Rose Luckin
Haha, yes!

Vimla Appadoo
And the reason I say this is with my, with my service design background, in that, I think if more businesses were able to implement in these artificial intelligence in a way that works, it would see us have seamless services across, seamless services and products that mean people's lives become easier. That it's easier to access things like universal credit, or access help and support across all sectors of society. And that's what I think the world will look like. I think it's one where people are able to let go of the stress of form filling or recruitment or getting a job because it's so much easier and simpler and done in a way that they know isn't disenfranchising them but empowering them to make the right decisions for them. And the knock on effect to the businesses is more business, better business and having a social impact.

Rose Luckin
That sounds good to me. Great. I really, really enjoyed this podcast. Thank you so much. That was a fascinating discussion. So thank you, to our guests. It really has been a huge pleasure to hear your insights, your thoughts. Fantastic! So thank you for sharing your expertise with us today. Our guests today were Vimla Appadoo, head of experience at Culture Shift and board member at the Greater Manchester local enterprise partnership, as well as the trustee for Diversity UK and Dana Elman Vishkin, innovation delivery manager at Digital Catapult and former policy advisor for innovation, enterprise and trade at the Federation of Small Businesses of the UK. Thank you!

Rose Luckin
You've been listening to working well with AI. This episode was presented by myself, Rose Luckin. Editing and mixing is by Suzie McCarthy. The series is funded by UCL Public Policy, UCL Grand Challenges and the British Academy. To find out more about the AI and the Future of Work Project for “UCL AI and the future of work”. Thanks for listening, and I hope you join us again next time.