Episode 10 – Integrating transport delivery with data and AI, with Chacasta Pritlove, Google Cloud and Katherine Williamson, Department for Transport
Epsiode intro
Our guests are Chacasta Pritlove Head of Transport, UK public sector, Google Cloud and Katherine Williamson, Chief Data Officer for the Department for Transport (DfT). In this episode Katherine and Chacasta talk about:
The unique data challenges the sector faces with infrastructure lasting 30-50 years, requiring organisations to balance long-term asset management with modern data needs while ensuring quality and accessibility.
The importance of having high quality, standardised data that’s been collected with its end use in mind.
Why data sharing and openness across transport modes is important and while it remains complex to tackle it’s still also very important to do.
How the focus is shifting toward making transport more accessible and equitable through data-driven solutions, with digital twins and AI helping to create more responsive and inclusive transport systems.
The need for behaviour change alongside systems and workflow changes to embed data and AI across organisations and society.
The Interchange Podcast is produced in association with Arcadis and interviews leading changemakers and thinkers about how we can make integrated transport infrastructure happen. This season’s discussions fall under 4 key themes: Place, Data and Digitisation, Energy and Environment.
The Interchange Podcast series is part of Interchange, which is a platform that culminates in a thought-provoking two-day major conference being held on 04/05 March 2025 at Manchester Central. Make sure you’ve registered for your free ticket at www.interchange-uk.com
Katherine Williamson
Katherine Williamson is the Chief Data Officer for the Department for Transport (DfT). Her role encompasses the Transport Data Strategy and the internal data programme looking at improving its data maturity, use of technology and skills.
Chacasta Pritlove
Chacasta Pritlove is the Head of Transport, UK Public Sector from Google Cloud. She is a passionate and creative thought-leader, driving digital transformation across transport, infrastructure and the wider industry - with 25 years’ experience in enterprise technology, and a focus on the public sector for the past 15. She is proud to have earned the role of trusted partner to several large organisations across the public sector which focus on integrated, seamless and sustainable transportation.
Resources and Links
About Interchange UK
The Interchange Podcast series is part of Interchange, which culminates around a two-day major event about rethinking transport infrastructure taking place in Manchester on 4/5 March 2025. If you’d like to attend you can book your place here.
Transcript
SPEAKERS
Chacasta Pritlove, Ayo Abbas, katherine Williamson
Ayo Abbas 00:03
Hello and welcome to the Interchange podcast, which is kindly produced in association with Arcadis. The podcast is hosted by me Ayo Abbas. I am an independent broadcaster and marketeer. Interchange is dedicated to transforming and modernising transport infrastructure networks through integration, decarbonisation and digitization. Each episode that we do of this podcast looks at how we can really make integrated transport delivery happen. My guests today are Chascasta pritlove, head of Transport UK, public sector at Google Cloud, and Katherine Williamson from the Department of Transport, who leads their data insights and people analytics division in today's episode, we're talking all about data and AI and its role in making integrated transport happen. We discuss the sector as it faces up to its unique data challenges with infrastructure that is, you know, lasting 30 to 50 years. And it's like, how do they balance their long term asset management with modern data requirements. That's a huge challenge for all infrastructure operators. We also talk about the importance of having high quality, standardized data that's been collected with its end use in mind, we share about data sharing and openness and its importance in transport modes and how while it remains complex, it's also something we really do have to tackle as an industry. We look at the shifting focus towards making transport more accessible and equitable and inclusive for data driven solutions. And finally, we also touch on behaviour change and systems and workflows and how we really need to make this change across organizations and society as a whole. Throughout the episode, we'll talk about use cases and practical applications where data and AI are already making a difference in transport too. So let's get on with listening to the interview. Hello and welcome to the latest episode of the Interchange podcast.Can you introduce introduce yourself please Chascasta,
Chacasta Pritlove 02:04
yeah, sure. Thanks. Ayo. So my name is Chacasta Pritlove. I work for Google Cloud, and I head up our public sector transport vertical. So what that means is that I work with the Department for Transport, hence I know Katherine, and I also work with all of their agencies and non departmental spending bodies, and I effectively stop when we come to local transport, where I hand off to my local government colleagues, but I will get brought in to, obviously, talk about transport related issues. So I'm here today to talk about the tech that enables the transport network to work.
Ayo Abbas 02:37
Fantastic. And Katherine, can you introduce yourself and your role at the DFT Police Department for Transport?
Katherine Williamson 02:43
Yes, of course. Thank you for having me. I'm Katherine Williamson. I am the Chief Data Officer for the Department for Transport. What that means is I work with the central government department to improve the way we store and manage and work with our data. Hence, working with Chacasta, because we use the Google tech as part of that process, and I also work outside the Department for Transport, so we support the transport sector as a whole. So we provide leadership through our transport data strategy and helping various parts of the transport sector think about how they should be working with data to get the best possible outcomes for the transport network.
Ayo Abbas 03:22
Wow, that's a really interesting it's a broad role, isn't it, but very it's very important at the moment, especially with how things are changing with transport and data. And so I guess my first question is, What do you mean by data and digital digitalization, which I can never say. Do you want to he wants to start with that one.
Katherine Williamson 03:38
So to me, data is everything. Data is everywhere. People are producing and using data all the time. But perhaps to illustrate how data is changing, talk about perhaps traffic counts fairly simple. We've kind of moved from a world where people are standing on street corners with clipboards and surveying as people come past, to having sensors in the road to being able to look at CCTV to then moving into using various other sources of data, like being able to see where people movements are or where traffic is the most dense through applications like Google. So the whole landscape has evolved from that kind of pen and paper tally count all the way through to the most modern technology. And for me, that's what makes it really exciting, is how we make the most of those opportunities as we move into this new world where, you know, everything is data.
Chacasta Pritlove 04:30
So for me and Katherine is right, and data is changing, but it's a it's a fundamental building block of everything, and in its purest form, data is raw facts, it's figures, it's numbers, it's text, it's images, the sensor readings that Catherine just mentioned. So it's that foundational building block for us to understand the world around us and alone. It might not mean anything and it you know, a single fact or a single figure can be interpreted in many ways, but you can. Combine data together and that gives it context and nuance. So you know, for us, it's that foundational building block for everything else,
Ayo Abbas 05:11
fantastic. And I guess in the past few years, AI is really kind of artificial intelligence has really come to the fore. So what role does that play in terms of data and digitalization. Do you want to kick us off Chacasta?
Chacasta Pritlove 05:23
AI is so I was reminded recently by Thomas ableman from Free Reeling of a quote from Mustafa Suleyman who was one of the founders of DeepMind, which is our AI research arm here at Google. And he said, AI is what computers can't do once they can do it becomes software, which is pretty accurate, I think. But to give a less flippant answer, so AI is the ability of computers to attempt to mimic human intelligence, such as learning, problem solving, decision augmentation or decision making, depending on the context. So it's about creating those machines that can perform tasks that typically require human intelligence. And a lot of the stuff that we're seeing is taking those really mundane jobs that we have to do of kind of trawling through data and actually giving that to effectively a robot to do so that we can concentrate on doing much more important stuff. So an example of AI that's actually been deployed in Greater Manchester is where Google research have used AI and Google Maps driving trends to model traffic patterns, and they're doing that, and they're building up intelligent recommendations for the city traffic engineers to optimize traffic flow in an effort to improve air quality. So they're using that. And early numbers are showing that, you know, 30% reduction in stops, 10% reduction in greenhouse gas emissions. So that's AI being used in transport in the UK and having fantastic results to show for it. So that's just one example. Then you've obviously got Gen AI, which is completely different kind of form of AI, and that's about creating new content. So it's about using those patterns that are unknown about and understood to create new content, like text or images or music or even code, learning those patterns from that existing data, and using that knowledge to generate new and original outputs. And I think a really good example of that from the transport side of things, is where we're deploying Gemini, which is our generative AI capability, into Google Earth, and that's allowing urban planners to access really deep city level insights and reduce the amount of time they spend analyzing data. So imagine you're a transport planner, and you want to understand where to put an EV charger. So you can now, in natural language, ask the system, can you map the five postcodes with the fewest EV chargers relative to their geographic area size. And it will use multi step reasoning, and it will give you back those five post codes. So again, that's a really, really useful idea, and you can then build on that to kind of take it further and ask if there are any hotels or shopping centers that don't have EV charges within a certain distance and that kind of thing. So it's really moving forward at pace at the moment, and you can basically do that like a click of a button, isn't it? It's passed as well, isn't it? Yeah, yeah. Whereas previously, you'd have to kind of trawl through all sorts of data yourself and kind of pull all the data together yourself, where, yeah. So it's completely changed the landscape.
Ayo Abbas 09:00
So Katherine, where are we currently, now, in terms of the landscape of using data and kind of transport as a sector? Are we ahead? Or what's the state of what's happening, in your opinion?
Katherine Williamson 09:11
So I think we have to be careful when we say transport as a sector, because actually it's lots of different sectors doing lots of different things. And you know what might work in Greater Manchester, as Chacasta just described, might not work in a smaller local authority like Rutland. It might not work for busses to say it works for trains, and that's one of the challenges and the opportunities in transport is actually, how do you get all this data to come together and work together, regardless of where you live or what mode of transport you want to use, and geography is a really important part of that, because in a lot of cases, it's that geography where something is that allows you to bring the data sets together, not just from within transport, but you might want to look at energy, you might want to look at climate. It. You might want to look at health, but how you bring all of those things together to tell us new stories, to provide new insight. But if that data isn't there, if it isn't of the right quality, if it can't be linked together, then all those fancy AI tools that Chacasta described might not give you the answer that you want all the answer that you need. So for me, getting those fundamentals right on the quality of data, the standards of data, the geography of the data, in a way that it can be linked together to allow all of these new insights to come about is really key.
Chacasta Pritlove 10:36
That data quality point, I think, is absolutely key. And I think that transport, because it's a long term game transport, right? So you people in transport are used to dealing with infrastructure that's going to be around for 30 to 50 years plus and and that's great, because we don't want to be building new roads and airports and train lines and everything every few years. So that long term view is essential. But from a data perspective, it's really difficult to deal in those kind of time frames, because, because data ages, and the data that you know is being collected on assets that are sort of 5 to 20 years old is very different to the data that we need now, and so the quality of the data that exists within transport does suffer, I think, from having that kind of long term mindset. And we do need to kind of have two ways of thinking about transport and Katherine's right. Obviously, I'm saying transport, but there are all the different elements that Katherine's described. So yeah, that long term view and that different modes, different areas, different locations, all kind of play into that as well.
Ayo Abbas 11:59
So what could we, I guess, change in terms of how we collect data in the in the sector to kind of, I guess, bring it up to the quality that would make it more useful, is that the right question, is there anything that we could change? Do you think Katherine or try to, kind of put in place to help that happen?
Katherine Williamson 12:15
So I think one of the things is, when we're collecting data, we have to think about what it could be used for above and beyond the purpose for which it's being collected. And actually, there's a small legal point here about actually making sure that when you're collecting data, you can then share it, and then you can then use it for other purposes. So thinking broader than the actual problem that you're trying to solve, I think there is definitely something about standards, making sure it can be linked with other data sets, be that on a geographic basis, be that on another basis, and then that quality point, that timeliness point, you know, years ago, it was fine to be able to know what your assets looks like with a big gap between them. But actually now, the way things are evolving, you need to think about things, not necessarily real time, but far, far quicker. How do you do that? How do you embed technology that will give you that information without necessarily a human being having to go out and survey and, you know, Network Rail is doing some really interesting things with all the technology on some of their trains, about, you know, can you scan the tracks to see whether or not they need work doing to them? Can you detect what's happening with the embankments so you know where you might want to intervene before there might be, you know, a mudslide or a tree come down. So all this technology is allowing us to move into a space where actually we might be able to predict and prevent things, rather than waiting for something to happen and then dealing with the consequences.
Ayo Abbas 13:50
That's brilliant. And so do you want to add to that Chacasta?
Chacasta Pritlove 13:53
The predicting thing. That is absolutely a use case that we're seeing again and again and again across the whole of the transport industry, both in terms of, you know, the embankments that and tracks that Katherine was talking about, but also in, you know, the engines of the of the trains in the assets that are along the roadside all the way through. So Asset Management and predictive maintenance is a huge use case and a huge benefit to be able to proactively maintain so that we're not disrupting the public when things break. And I live on the hook line, so I'm quite happy if there was never going to be a landslide on that side of things again. So that that would be fantastic personally. But I think also to pick up on Katherine's point around the kind of what are you collecting data for? I think that's that's a really interesting point that we have to be really, really, really, really clear about. So. From the Google side of things. When we collect data, we have to be really, really rigorous about what that data is going to be, and we have policies and stuff about what is that data going to be used for, and we don't collect it in a well, we might be able to use it for this, or we might be able to use it to that it has to be specific, because there are rules and regulations that we have to adhere to to be able to do that. And you know, if we're just collecting it, just because, well, it might come in useful one day that's that's not on, right? So we have very, very strict rules about when we're designing products, about how do you envisage that data being used? How long do you need to keep it for? How exactly are you going to use it? And that is something that, I think, from the transport industry perspective, that to my understanding, I haven't seen, certainly, any centralised kind of view of kind of how or guidance really is to how that how that should be done and how that should be worked. And I think especially now that AI is coming into that and the general public are more aware of data and AI and how their data is being used, that there has to be that kind of openness from the industry as a whole. And I'm not just talking about the department and the government, I'm talking about the industry as a whole, about how all of that is being brought to bear and what is done with it.
Ayo Abbas 16:40
I mean, they're very I guess it's those upfront opening conversations, isn't it, about this is where we want, what we want to do, and being absolutely clear about that, isn't it? And I guess those are conversations we've not necessarily had before. We've not we've not been in a situation where we've needed to have them. But you're absolutely right. It does make sense. And in terms of if we want to get to a world where, I guess data sharing across kind of modes, nodes is happening in organizations. I mean, what changes do you think you would need to happen to make that more of a more of a more common place? Katherine,
Katherine Williamson 17:11
so I think there are, there are a couple of aspects. There are lots of different barriers at the moment, and every use case is different, and you need to understand those. But I think a lot of it comes down to conversation and collaboration. So and a lot of times, the people that will perhaps benefit from the data being shared aren't the people that currently own it, and therefore the people that own it can't necessarily see why they would want to give that data up. So, I think a lot of it is about bringing people together to have those conversations, getting the people that own the data in the same room as the people who might be wanting to undertake the analysis, and potentially some of the people that will benefit from that analysis once it's done. So, you can bring everyone together once you have those strong use cases, actually the barriers to sharing data become a little bit easier to overcome because everyone's working towards a shared, common goal. So, you know, there is regulation in place around data sharing. The technology facilitates data sharing in a way that it's never done previously. You know, in some respects, it's very, very easy to share huge amounts of data now technologically speaking, but to overcome, overcome the barriers about people being reticent to do that. It really comes down to those use cases. Why, who? How, all of that needs to be done in collaboration. I think for you to really have those successful examples.
Ayo Abbas 18:41
And I'm guessing one of the things the UK case is how it will, how that will impact me as a user, right? As a transport user, or as a member of the public, that's quite important to bring those, those stories to the fore, right?
Katherine Williamson 18:52
Yeah. And you know that transparency, that ethical piece, is really important, particularly if you're going to a place where you're applying AI. AI means different things to different people, and being able to explain to people actually, this is the problem we're trying to solve. This is the technology we're using to do it. This is how we're storing your data safely and securely as we do this. And these are how the outputs will be used. Is a really important part of that process of making sure that everyone in the sector and the public understands and trusts why we're doing some of the things we're doing in this space.
Ayo Abbas 19:30
Anything you want to add to that Chacasta?
Chacasta Pritlove 19:31
I think the opportunities are fast, right for both the passengers, the freight users, for the public agencies that whole we've touched on the kind of the predictive maintenance and the the ability to reduce the pain that the traveling public sometimes feel, but it's about giving them that kind of personalized journey planning. Those kind of seamless, multimodal journeys. Because, you know, we all, yeah, I don't know anybody that just uses one form of transport to get to where they're going. You have to use all of those different so, you know, I drive to the station, I have an EV, I drive to the station, I get a train, I then walk. So it's bringing all of that together so that I can actually do all of that without sort of having to be inconvenienced and having multiple apps and all of that kind of side of things. But it's also, you know, about being able to operate that that network better, and, you know, optimizing traffic flow, reducing congestion, improving the scheduling so that I can do that multi modal journey, and I'm not going to, you know, miss my train because my other train has been delayed, or my bus has been delayed, that kind of thing. I think there are some challenges that we need to to overcome to achieve that. I think that currently individual agencies and operators Katherine said, they don't always own the data that they need, but when they do own the data, often it's in legacy systems that are really difficult to get that data out of, and it's difficult to then kind of combine that data sets with other data sets, because, as I mentioned at the start, just having one view is very subjective and open to interpretation, bringing all of that data together to give that that wider landscape, that wider view provides it with context, and you Can't do that when you've got these older systems that don't necessarily talk very well to the newer systems. So kind of that kind of standardisation and achieving that interoperability is something that really needs to needs to happen.
Ayo Abbas 20:11
I guess, in terms of like data and how bringing it all together. I mean, what kind of risks are there in terms of operators and suppliers and in this kind of area, what do they need to be considering? are there any particular things?
Katherine Williamson 22:09
So I think it depends what the data is. You know, the risk level changes depending on whether you're talking about information that's effectively already in the public domain, like some infrastructure through to if you've actually got personal data on individuals, that's a very different thing, and that, you know, there's laws and regulations GDPR around how you should protect that kind of data, as Chacasta says at the beginning. You know, you've got to think about what you're going to do with that data, why you're going to use it, and how long you're going to keep it for we shouldn't just be piling everything into a bucket, just in case it might be useful at some point. You really have to have a concrete use case, a reason, as I said, As for sharing it. And that will help you unlock actually, what data needs to be shared, how secure does it need to be? Who is going to be in control of it once it is shared, how it's going to be used. So there are lots of things for people to think about, but there are, you know, there are things out there to help. So, you know, we've done some work on data ethics with the Open Data Institute. They've got a load of resources, free resources on their website, as well as training courses to help people think about some of the ethics of considerations around sharing data. As I said, there's legislation about what you should be doing to protect personal data, and the technology will help you. We're working with Google at the moment, but you use Google Cloud, you can build things within the technology that will restrict access to certain fields so only the right people see the right information. So you can open up data sets safely and securely, and then protect some aspects of that data set as well. So there are lots of different things, and it is complicated, but as the technology moves and as people's understanding improve, there's a huge amount of resources out there to help people as well.
Ayo Abbas 24:02
Brilliant. No, thank you for that. It's really, it's really interesting. It's such a it's a complex thing to look at. But, yeah, right, it's, I guess, looking at different levels, isn't it? In terms of how you kind of store data and use it, that's really interesting. Um, and in terms of kind of AI applications, where are they kind of being used in transport at the moment? Really well. If you've got new examples, Katherine, or things that you're like, that's really interesting and new ways of working and new ways of using data,
Katherine Williamson 24:27
yeah, there's, you know, there's lots of innovation going on in the transport sector. So I've already touched on, sort of Network Rail example. There's other things that we're doing. So, you know, there's lots of work being done at the moment around digital twins. So that's a really interesting place where you're pulling lots of different data sets together to try and get this this model effectively of the real world that you can interact with. And that's being done in this country, but it's also been done successfully overseas, around ports, around road infrastructure.Those kind of applications there. The other things AI can do is you can do what humans do, as Chacasta said, but quicker. So it can do things like fraud detection, so you can sift through loads and loads of images quickly. And you can program that to look at, you know, where it might be unusual, why there might be things that you're concerned about. And then you can pass that to a human to make that decision. So successful, AI often has the human in the loop. Often has that interaction with a human. What it's doing is AI just enables you to do things quicker or more efficiently. It doesn't necessarily completely remove the human from the process, because, there are lots of reasons why you still want a human being to look things over.
Ayo Abbas 25:42
Fantastic. Have you got any examples Chacasta that you wanted to share? Or,
Chacasta Pritlove 25:45
yeah, absolutely. And I think with the advent of large language models, which are the basis of generative AI, so as an example, we are talking to some customers, where they are looking retrospectively at incidents that have occurred on the railway line. And there are huge amounts of incidents that are caused by trespass, and it as soon as trespass occurs, things quite rightly have to be really locked down so trains are slowed or stopped, and obviously that's really inconvenient for the train companies and the traveling public. So one of the things that they have to do now is they have to sit there and listen to the to the radios of the train drivers and try and kind of retrospectively work out how and when things happened, and so they can identify where the trespass occurred. They can go back and see if they can remediate. And they do that by somebody sitting there and listening to the transcript or reading the transcript. And so one of the things that we're talking to them about is, okay, well, if you put that audio file in a large language model like Gemini, you can then ask that audio file to take you to the place where the train driver says there's somebody on the track, or I've seen a person, and it will immediately take you there, rather than you spending 24 hours kind of listening to this thing. You can just get the machine to do it for you, and then you go to that place and you check that that's the right element. So I think that is that time saving is just humongous, and it will have such a great impact on the efficiency of organizations to be able to do those kind of things, and they can then kind of focus on making everything else better and doing that remediation and getting it done quicker, which, at the end of the day, gives all of us a much better traveling experience. And I think that's that's absolutely vital.
Ayo Abbas 27:56
There's a lot of this around building use cases, like looking at your current processes and understanding actually, this is a, there's a tool there that could help us to do that. Is that a lot of what this is,
Chacasta Pritlove 28:07
a lot of it is, it is about looking at those, looking at those workflows, and saying, Okay, so in Google, we talk about toil, and it's toil is that really boring stuff, monotonous stuff that we have to do, and looking to identify where that toil occurs, and how can we then use technology to remove that toil? Because when you remove that toil, you've then freed up however many hours, days, weeks, months, to be able to do something better. To look at improving the process is just look at making changes, rather than having to sit there in front of a screen or, you know, with your headphones in listening to audio files. So that is absolutely key. Technology is an enabler, and the only way that technology is going to to work is if it's embedded in the workflows. You can buy as much technology as you like, but unless it's embedded in your workflow, it's not going to change, it's not going to move the dial. So it's about the technology, yes, but it's also about the people, so training them up and making sure they understand how it's being used, what it's being used for, everything that we've spoken about just now. But it's also about the processes and how you build the technology to support and augment the processes that you already have
Ayo Abbas 29:30
fantastic and I guess that takes you on to that whole kind of behavioral change and communications and change management and change communications, which is a slightly bigger beast, isn't it, around bringing in new technology and getting it adopted and embedded in large organizations. So I say, anything that the DFT is doing around that Katherine, in terms of embedding kind of technology and data and AI and all of those,
Katherine Williamson 29:52
yeah, there's a number of things we're doing. I mean, for us, a lot of it is just those building blocks. Needed them to be right. So as Chacasta said, you know, legacy systems are still a challenge. We still have some of those within DFT, getting all of our data into modern technology so you can apply these techniques is a really important step for us. And then there's that, you know, upskilling the business, not just the analysts that will necessarily be working with the data. You know, doing training courses for them, yes, but whilst they're starting to think about how you widen the options out to people away from that community. So can you create more dashboards and visualisations to allow people to self serve for data and understand without necessarily going to their analysts to answer questions? Can you encourage people to be more data curious so they start to ask their stakeholders more questions about data, you know, bringing that in and then ethics training, you know, getting people to realize, yes, the technology is there. We can do this. Just take a moment and think, should we do this? Is it our role? You know, who are the people that need to be involved? How do we make sure we do this safely and securely. So all of that kind of comes together around the people element, I think, and doing that is a really important but it's a really interesting part of the role. I think lots of people get very excited by what the technology will do. That technology will only do that if actually people are willing to use that technology and know how to and want to use that technology? Well,
Ayo Abbas 31:23
no, that's really, really interesting. I was just going to say, didn't you? You did a hackathon, didn't you together in May last year that was bringing together lots of kind of different minds and ideas that must have been quite inspiring.
Katherine Williamson 31:34
Well, that's always great, and we love that hackathon.
Chacasta Pritlove 31:37
It was fantastic, absolutely brilliant.
Ayo Abbas 31:40
And what sort of ideas did you see from that? Lots
Katherine Williamson 31:43
of different ideas, interesting. A large number of the ideas were around that kind of accessibility point. So really starting to think about, you know, not just saying A to B, but actually how people want to get to a to b, and thinking around about things like, you know, curb heights or tactile paving, or where street lighting is, all these infrastructure pieces that come together to enable people to potentially have more accessible journeys, because there's a large part of people for whom the transport sector now doesn't currently work well. And actually, I think part of doing this in a fair and equitable way is, actually, how do we make the transport sector work well for everybody? An accessibility piece is really, really important.
Ayo Abbas 32:29
And I guess one of my final questions is, I guess, when it comes to kind of data and AI, so I know, for example, the government recently did an announcement that they're going to be supporting more kind of AI initiatives and things like that. So what do you think the future? What future is possible with AI and transport? Do you think I'm going to throw that to you? Chacasta?
Chacasta Pritlove 32:48
oh, gosh, that's a big question, a big question. It's a big question. So I think a step back first to kind of recognise, really, that AI is is not the panacea. It's not going to solve, you know, we're not going to install AI and the transport network is going to be fantastic. It's just not going to work like that. So it's, it's for me at the fundamental building blocks of AI. Right back to the beginning of this conversation is about the data, and we need to get the data right before anything can happen. So it's got to be good quality, it's got to be accessible, it's got to be secured, it's got to be used ethically, all of that kind of stuff. If we get that right, then we can create a transport system that is more equitable, is more accessible, is serving the people when they want it, where they want it. And you know, that's all really kind of big, big, fluffy stuff, and it's going to take a lot to get there. But, you know, ultimately, data, AI and transport. It should be used to benefit society, and it should be used to improve people's lives. And Catherine mentioned it at the beginning that transport is not an industry that is siloed by itself. It involves healthcare, it involves education, it involves the Environment, Energy companies, all of those. And I think there are benefits to all of those other industries by making sure that transport is the very best they can be. And AI is one tool that will help you get there. There are a multitude of others as well, and I think we need to recognize that, and kind of not just think of it as a big pill, but we can swallow and everything's going to make everything better. So an Ayo Pill is not going to work. Katherine, is there anything you wanted to add to that?
Katherine Williamson 34:55
Just that, I'd really reiterate that point around collaboration, there's some really good things being. Done, but actually, to get the benefit from all of this, it's about bringing people together. It's about sharing expertise. It's about sharing things that have worked, importantly, sharing things that haven't worked. We're not very good about saying, you know, the whole, the whole mantra about fail fast, fail cheap. Well, that means, you know, being honest when you've tried something and it hasn't worked. So for me, that bringing people together, from government technology, from the SME sector, from the charity sector, academics, that collaboration piece, I think, is how we will get this technology to really work to the best of its ability.
Ayo Abbas 35:33
Brilliant. Thank you both for coming on to the show. Everyone really enjoyed the conversation. Thank you. Thank you. It's been wonderful. Thank you. Thanks so much for listening, and don't forget, Interchange isn't just a podcast. It's also a two day conference that will be taking place at Manchester Central on the fourth and fifth of March 2025 The conference will bring together the key public and private sector infrastructure operators and their value chain to talk about new ideas to challenge the status quo and build stronger relationships so we can have a shared approach to make integrated transport happen. Head to www.interchange-uk.com to find out more.
