I'm going to replace my design team with AI, unless you tell me otherwise...

Episode transcript

Amelia (00:02.273): Hey there and welcome to episode 2 of the podcast. I'm Amelia.

Paul Wood (00:07.946): I'm Paul.

Russell (00:09.73): I'm Russell.

Pat (00:13.073): I'm Pat.

Beth (00:14.481): I'm Beth.

Mark (00:16.893): I'm Mark.

Amelia (00:18.387): This is Fin the Week. So, a big group today — how's everybody doing?

Paul Wood (00:25.706): Very well, thank you, yeah.

Mark (00:26.791): Thank you.

Beth (00:27.633): Thank you, thanks.

Pat (00:27.89): Kids?

Russell (00:28.439): Okay, thanks.

Amelia (00:29.645): Good, good. And some new faces as well, which is lovely. So last week, episode one, we started talking about AI, and of course we're building on that today. We've got a controversial statement to kick around this week: "I'm going to replace my design team with AI." Who came up with that — was that you, Paul?

Paul Wood (00:30.22): Full house.

Paul Wood (00:51.39): Yes, yeah, it was. It's because the whole industry is talking — the whole world feels like it's talking — about jobs that are going to be replaced by AI. And design is a particularly interesting one, I think, because, well, first of all, Russell raised the point in the last episode: what is design? And that got me thinking — actually, yeah, design is a lot more than what I imagined when I thought about that. When I thought of that question, I thought about

the type of design we do, which is interface design and UX design. But actually, then after that, I watched a review of a car that was designed by AI, and the idea was that no part of the material goes to waste, because AI was able to create these organic structures that built the car and made it really efficient. I thought, oh yeah, that's design and that's replacing humans. But yeah, I thought it'd be a good kind of controversial topic to discuss really, and see

what our design experts on the team think about that.

Amelia (01:52.184): We've got Beth, we've got Mark, both with design backgrounds, and of course we've got Russ as well. So what's your knee-jerk reaction to that statement?

Mark (02:06.226): Who's going to start?

Russell (02:06.286): Should I go first? It's definitely very topical at the moment, and the design process is changing rapidly almost every day because of AI. There are new processes that we're almost forced to adopt. And I think...

Russell (02:33.218): But before I explain some of those, I think the essence of design kind of remains the same. It's always kind of difficult, I guess, having been to design school — I was at design school for probably four years in Bristol. It's always difficult, I think, going into a commercial environment with

other people who probably don't understand the essence of what design is and what it means and what it exists for. So you're kind of constantly battling, almost, with the entire organisation to explain what design is, how important it is, and how it can deliver some amazing results. Almost to the point where,

if designers aren't heard, they kind of just become operators, they kind of just become production staff, they kind of just become the ones you hand a wireframe to or a scope and say, "Can you make it pretty?" or "Can you jazz it up a bit?" — those are the famous words. And

designers in organisations that I've seen often just become implementers of other people's ideas, and other people's problems that they've solved, to make them look good. And that isn't design. That's really a fundamental shift that, if you get right in an organisation, can have a massive impact. Design, the way I see it and the meaning of design, is: design exists to solve a problem,

or meet a need. It's that initial need or problem, which is usually set by the client at high level. It could be set perhaps by a team — a marketing team, for example, might need something — although it's probably less of a problem to solve and more of a production role in that case. But certainly, the way I've been schooled about design — at uni, for example, which is a long time ago, probably

Russell (04:40.11): 25 years ago now — I was at uni in the early 2000s. We were never asked to design anything. We were always asked to solve a problem. We were always given a brief, and it was like, here's your brief, come up with an idea which

solves a need. And then you decide what kind of method to use, you piece it all together, and as part of your exploration work at the start, you'd create a hypothesis to test and see if it solves that need. So there's kind of two different types of designer, really. The designers who just do production —

I think you'll see those worlds kind of disappear because of AI, because you can start things so quickly. You can put in

frameworks and a lot of design theory into things like Markdown and use Claude, and it can deliver designs. But ultimately, that's just a production method to solve the problem. And it's that initial problem, and solving it and the output, that is and still remains the design process. It's almost quite hard to explain, so I'm kind of bumbling my way through it a bit, but

I used the example of a chair last week and I feel like I need to see that through, because "how do you design a chair" is a really interesting one — everyone's probably sat on a chair, so you can relate. But designing a chair, you think about, the problem to solve is: we need an end result, which is something that someone can sit on, right? But

Russell (06:41.664): at the beginning of that process, you need to ask a number of questions. You need to say: who's using it? Do they have any access needs — for example, accessibility needs that we need to take into account specifically for this chair? What kind of environment is this chair going to live in? Is it going to be hot? Is it going to be cold? Is it going to be in someone's lounge? Is it going to be something they can just chill in? Does it need to fold up? Is it going to be at the beach?

There are loads of questions you need to ask in order to design something which seems as simple as a chair. And then there's the commercial viability of that chair — what materials should it use? What kind of trade-offs do we need to make in the production process to meet those business requirements? All of this creates a kind of hypothesis, which you test when you produce that chair. And you test it with the people who are going to sit on it. You wouldn't use AI to come up with an idea for a chair and manufacture the chair

and then put it in all the shops, because that's probably going to go disastrously wrong. I think it's quite a good example, because if you think about designing a chair and getting a designer involved, the opinion might be that the designer just thinks about what colour it is and perhaps what material to use. But actually, the designer comes up with that whole problem to solve.

And that's the school of design, and that's kind of the design thinking — that designers, in my opinion, should be higher up in the chain when it comes to not being given a brief on what to do, but just being given a problem to solve.

Think about how that solution aligns with business objectives, how it's commercially viable, how it can be produced, and how we can test it and validate it before it goes to market — and then continue to test it, validate it and measure it until we get it right. That's how I see the design process.

Russell (08:44.814): And AI obviously comes in and speeds up production. It can help with idea generation. It can do loads of stuff. But I can't see it ever really taking away that initial step where someone thinks about what to do to solve a problem, and then also perhaps becomes accountable for that, and tests it and validates it at the end. I've used the example of a chair, but you could apply that entire process to a website, an app,

anything — anything you can think of, a car. But that's, high level, how I see design, I think. I'd be worried if I was just a production designer — like a Figma operator, that kind of designer. You definitely need to be thinking strategically, and you need to be experimenting with all these tools as well. But

I don't think AI has changed the design process — that traditional design process. Just one final thing, because I could talk about this for an hour. The process that has been introduced, which is really interesting, is supporting and consulting on things that are AI-generated. So there's that traditional process where you do the research,

you produce something and you test it. I think that will remain. But the interesting process we're seeing now is, you get engineers and developers with their army of Claudes and they're going to go away and produce something which looks really amazing, to be honest — a really amazing interface.

We don't want to stop that process as designers. We don't want to say developers, or marketeers, or project managers shouldn't be exploring, designing interfaces and creating them. The process that we're being forced to adapt to is just to allow that process to happen, and then the designers come in as more of a consultant and basically provide some guidance on

Russell (10:53.722): why they think things should go in certain places in interfaces, based on their experience; how to introduce some success metrics to see if this work is going to have an impact and meet those business objectives. It's a completely new process that we are adapting to. But I certainly wouldn't want to hold up that new process that AI is bringing, where other people can design interfaces. And then you have the more traditional

discovery, research, do some mock-ups — with the designers now also being able to produce quite advanced prototypes using AI. So yeah, it's a bit messy at the moment. We're kind of figuring out the best way to do things, and it will almost certainly be per project, depending on who's working on it and what the project objectives are. It's a really exciting time, because

so many people can get involved in that creation of websites and apps now — more than ever before.

Russell (11:59.587): I'll end it there for now.

Amelia (11:59.735): Sorry. There's a lot to unpick there. Has anyone got any thoughts on what Russ has said?

Pat (12:08.965): I've got some thoughts — just waiting for us to get some feedback from Mark as well. It'll be really useful for you all. Big change really is in the tools. So the tools out there allow

Pat (12:29.293): you to produce good-looking layouts, and that's the key word there: good-looking. Not necessarily well-functioning — good-looking layouts. These tools allow, if you're a developer and you've already got any design experience, you can just go into Claude and say "build me an interface" and it'll come up with an interface that, at first glance, looks really nice.

Pat (12:56.283): And that ability to create a nice interface is something that traditionally a designer would have had to have done, but AI does it now. However, what AI can't do is make that interface function really well, in addition to looking nice. The devil's in the detail when it comes to really good application design, interface design. It's about making sure that

the validation appears in the right place, the user journeys are bang on, and everything's consistent. From a technical perspective, that human input — and having experience in design and producing those interfaces that work really well as well as looking good — you definitely still need a human for that, and I think you probably will for some time. So that's just my feedback there, yeah.

Amelia (13:56.649): Mark, what are your thoughts?

Mark (14:00.006): There's an awful lot to unpack with it. Obviously, in the current situation we're in, it's very new, and everyone's trying to find their feet with it in terms of adopting these new practices into a workflow or a process or anything like that.

So I think there are challenges, even at our business scale, but obviously industry-wide as well. What Paul was talking about at the start — it'd be interesting to see how AI is influencing product design from that perspective.

But from my perspective, working in UX and UI design, I believe AI will make the collaborative process a lot easier. I think it will bring people together at the start — engineers, consultants, stakeholders,

designers — can all go in a room and flesh out ideas. Where AI brings the advantage at that point is, obviously, the speed: it can bring things together and you can get initial solutions out quite rapidly. Those solutions aren't necessarily correct, and then the experienced human side of things can

apply your experience further down the line to make these solutions correct. But essentially it can make people collaborate a lot quicker. In terms of graphic design, that's where I see the challenges. I think AI is making design a lot more accessible to people,

Mark (16:01.184): and I'm seeing an awful lot of "I can do this" or "I've done this", and it's very much about the eye. Those people who are using it in that way are not thinking about the UX things — in terms of who, what the problem is

to be solved? Who are you solving that problem for? You see what I mean? Those kinds of requirements. So those rapid solutions — they may work for a bit, they're going to fail because of the lack of thinking around these types of things. But in my own processes,

I am starting to integrate it into what I do with designs. It can allow you to validate solutions really quite quickly, and that's really useful, especially when you're working with complex systems and stuff like that. You can test things, you can get solutions in front of users, they can test it, they can give feedback — it makes that

iterative process a lot easier. So at the moment I think it's great. I think it's a real positive thing. It's just that adaption, and making everything kind of streamline with it. That's my thoughts.

Amelia (17:31.756): And you touched there upon — you mentioned graphic design. According to the World Economic Forum's Future Jobs 25 report, graphic design is number 11 on the list of fastest-declining jobs. So this is happening, isn't it?

Mark (17:45.692): Yeah, I know. I found that a mind-blowing stat. What they're talking about there is the rapid production of things like social media posts, the ability to produce... I always think the graphic design skills are with typography and creation and branding, and

applying those skills there in those upfront tasks. So I feel there's always going to be a need for that. I know a few illustrators as well — they're the ones who are suffering, because it's so easy to generate that style of imagery at the moment, or specify a style of imagery. So that's difficult.

In that line of work, people are going away, creating illustrations, writing storybooks, publishing it and doing it within a day, which seems madness. I think speed leads to a lot less quality. I think the people that will stand out are the ones that focus on quality.

Russell (19:04.878): Just to add to that — you lose a lot of authenticity as well. Big brands — I've seen this, I can't remember who it was, but they basically put a "human-made" kind of stamp on this advertising campaign that they did, and it was all illustrated

by a professional illustrator who they actually highlighted, their profile as part of the campaign, who did it. That's going to be much more powerful for the big brands that can afford to do it than using AI to quickly manufacture some of this work. So I think there are going to be different cases, aren't there, for using it.

Russell (19:51.887): A small agency like ours, without any professional illustrators — it could be a really handy feather in our cap, in terms of a style of graphics that we can produce quite easily. But we certainly wouldn't position ourselves as expert illustrators, and we would let the client know if we, you know,

I would let the client know if I'd used AI to produce an amazing illustration, because of a couple of reasons. From a licensing aspect, that's really important to think about as well. If you produce anything, do you have a commercial licence to use it in a campaign? You need to think about that. So you need to be upfront about the source of that artwork.

But yeah, I think "human-made", "handmade" stamps are going to be quite powerful — quite a powerful thing to be putting on things like illustration and music and that side of it.

Amelia (20:56.16): That's interesting you touched upon that, Russ, because that was actually something I was going to ask. When it comes to your clients, how transparent do you have to be? And how much do they care, if they've got a good end product? Do you find that clients really want to know step by step how you got to that final product? Is it something that you are seeing more of? How does that kind of work?

Russell (21:21.528): Well, from a graphic design perspective, it's just when it comes down to licensing, really, and how it's made. I just see AI as a production tool. If we use AI to create a prototype or write some code, or

it certainly isn't — it's just part of the process now, and that's happened quite quickly. So how we produce it is kind of up to us. It's just part of the steps that we take to produce something. And the way I see it,

I'm not a developer, but we used to use Stack Overflow quite a bit, kind of copy and paste a lot of code from there. We certainly wouldn't explain that whole process to a client. The same way, we wouldn't explain using AI. We use icon sets in design, we use stock photography,

stock kind of background creative, and often it's free and just becomes part of our process. We wouldn't always explain exactly where all of that came from, especially if it gives us ideas to build upon.

Paul Wood (22:38.633): The thing I found interesting from the World Economic Forum report that we looked at was: graphic design was one of the fastest-declining jobs, but UI and UX design were the eighth most in demand. So they're obviously growing in demand. Does that tie back to what you were saying earlier, Russell — that design isn't making things look nice, it's solving a problem? And

actually, could you see it that UI and UX designers are the problem-solvers, maybe, and the graphic design element is perhaps more at risk from AI because it's... I'm massively oversimplifying this here, but it's putting the paint on, isn't it, and making things graphically look good — and that's where AI can assist. Is that a fair assessment?

Russell (23:34.319): I think there are a couple of things going on with that report, because I also saw Autodesk analysed 3 million job listings, AI job listings, in the last couple of years, and design skills came out on top as the most in-demand skill. Technical and coding and all of those were beneath that, which was a real surprise to me. And it's quite

quite relevant to the topic of this podcast, but I think a couple of things are happening with graphic design. There are a lot of subsets of graphic design. Going back to when I was at uni again — I studied graphic design, but on my course I specialised in all the interactive media modules. I chose a route of interactive media on that course, and I was doing a lot of ActionScripting and 3D animation. My final piece was this

3D environment where you could click on different elements and they would respond and represent different sounds. I was going to take that piece of work and that research and do an MA, but I was just really keen to get out of uni and start earning some money, to be honest. But that was all under graphic design.

But essentially it was UI design, thinking about user experience. There were other people doing graphic design who were quite heavily on the print side of things — they were doing screen printing, they were really obsessed by typography. It's such a broad umbrella. And I think probably what's happened, as well as the decline in that job role, is you probably don't get as many traditional graphic designers as you used to. People are much more specialised now. They're specialised in UX,

in UI. Perhaps 10, 20 years ago — or 10 years ago — those people might have labelled themselves as graphic designers, but been doing a kind of UX/UI role.

Russell (25:32.687): So that might be part of why you're seeing this change in job roles, and some — what you could say is — a subset of graphic design moving up that list, because fewer people are labelling themselves as graphic designers. Yeah, I do think, from a UX perspective, the reason why it's rising in demand — I think

because, alongside AI, how do you measure the output of AI-generated work when it comes to interface design? And how do you create evidence-based AI work which meets the objectives of a business? Ultimately it comes back to ROI. So if you have a business which employs

a production team, you don't want to risk investing all that money into the production team without making sure that you can measure the outcome of what you've done — that there is some research, and that you have someone or a team of people involved to understand the user. The difference I see a lot with designers and perhaps engineers is, designers naturally have that empathy and that understanding towards the end user, which is quite a difficult skill to learn.

And that's probably why these people perhaps became designers and UX designers in the first place — because it's quite a difficult skill to learn. But I think those softer human skills are what makes this whole AI production process work and what makes it valuable — and I think gives the best possible ROI for that business investing in it.

Paul Wood (27:19.173): I was going to ask — come back to the point we discussed about, do clients care about how you reach the result? Beth, from your experience — because you sort of have an interest in design, but you're also a project manager — in your experience, during a project, are clients desperate to know the detail about how we achieve what we achieve, or

are they purely focused on the end result? And does AI change that?

Beth (27:54.479): I think it's all moving so rapidly that it's still quite a new conversation, and it's not something that we've encountered much, actually, so far. As we go on, it will crop up more and more. And I think there'll be — like we discussed recently internally — where does the ethics debate come into this? Should we be disclosing what we're using AI to support us with, and how much of our output is supported by that? And I think it comes back to that whole point we've been discussing around

output versus outcomes. You've got the how and the what. AI is really good at that if you want something to be churned out quickly. But thinking about the outcomes and the why — that's the important bit. I suppose that's where we've got this overlap of, what do we disclose, what have we done? And we haven't really encountered that yet. So I think it's something that will definitely crop up more in the future. And it's moving so quickly at the moment that I don't think,

if there were any guidelines or legislation, they're probably not able to keep up with the pace at which this is all changing. I think we've all really noticed it since the new year — I have, definitely, talking to the team. Previously we'd have had calls where, perhaps, we're a bit behind schedule and we might need to talk to the client about how we implement something. And actually now it's so much quicker. There are so many more ways that we can do it. And perhaps other team members that couldn't previously

pick up work or a specific task are now picking up those tasks. So it's changing every aspect of how we work. Yeah, I think that remains a bit of an unknown for now, Paul, and I think it will depend on if any guidelines or legislation come in. Perhaps we need to look internally at how we — what's our own rule book or guidelines — and are we upfront with our clients about how we use it? That might be a good way to lead on that.

Pat (29:50.142): If I can just, from a coding perspective — we're using it.

Pat (29:59.667): Largely, our clients are just really happy with the work we're doing, because we're able to deliver more features, we're able to turn work around much

Pat (30:10.418): We don't have to disclose that we're using AI, because, like you said, Russell, earlier — it's just a tool. It's another tool that we're using to just do work more efficiently and to a higher degree of quality. I was really excited actually yesterday — we were talking through a proposal to solve a problem that they were having, and I was actually really chuffed. It's a nice hand to have:

"We've got a solution to your problem here, and we're using AI to deliver this solution, and actually, because we're using AI, it's going to cost significantly less than

if we'd have pitched this two or three years ago." And as such, as a result, what we're proposing now is a really compelling solution to the problem we're facing. Whereas two or three years ago, the cost would have been prohibitive to roll out that solution. As part of that pitch, I was like, we wouldn't usually go out to really solve this problem.

Pat (31:17.466): Received very positively.

Pat (31:31.775): Assumptions that they just kind of assume we are. Maybe they don't. But honestly, if a client said to us, "You're not allowed to use AI", we would genuinely question whether it's worth us taking on that client, because it would feel quite backwards going back to how it used to work — because it's so powerful, and it really is allowing us to do the work.

Mark (32:10.472): I guess that's where the importance of tracking and understanding the goals you set with a client, and then the outcomes at the end, are the important parts. The time, the mechanism for delivery that happens in the middle — have they ever really cared about that? If you're talking about setting up CMSs and stuff, that gives you something to focus on in terms of selling. Well, these days it's much more

establishing those requirements to deliver the outcomes at the end. And that's where the quality will show through. And that ability to engage in that process, with users,

with the tools to show clients that you are delivering those outcomes — that is going to be really important for agencies going forward.

Amelia (33:16.204): And I suppose that's the whole relationship thing with your clients — just having that trust in you to know that you are going to create the best outcome with the tools that you are experts in.

Mark (33:30.066): Yeah, absolutely. It's always about building a relationship with a client so you can have that back and forth. Obviously they want a product delivered to a certain time, and with pretty much 90% of the projects, it works that way.

But then it's the relationship afterwards in terms of tracking those metrics, making sure that they're happy, extending the functionality or reducing it, dependent on how a user engages with it. That's the interesting part — what happens afterwards in terms of understanding user behaviour, especially within the types of products that we build.

Paul Wood (34:18.888): When I was looking at the car example and how they used AI in that, the brief for the generative AI would have been to basically avoid any form of waste, and create elements of the vehicle that are as light as they can possibly be and as strong as they can possibly be. I was thinking about, in the type of UI design we do and what our clients are interested in,

I suppose the key thing is to make it as usable as it can possibly be and to comply with regulations. Consumer Duty is a big topic of discussion in the financial services sector, and making sure that an interface, or a piece of communication, can be understood by the end user. It's not just a process they can work through; they actually need to be seen to be understanding it. Is there a place for AI to

help us in validating that sort of thing? Because when you're designing a physical product like a car, you can literally validate it — you can weigh the car and you can test it. Can you use AI in a similar way in a more knowledge-based sector?

Mark (35:38.152): There are usability principles and accessibility and all these things that relate to Consumer Duty

that you could almost train AI on, I suspect — you can plug in and then validate any solution, whether it's AI-generated or human-generated. If it can assist in making that side of things much more efficient,

then that again just helps the process and helps deliver on the outcomes in the right way. It's training that and using it in that advanced way. I'm not seeing any of that yet, but it'd be really interesting to explore.

Russell (36:35.406): It's kind of two-pronged, isn't it? It's using those frameworks to review — or audit, as we sometimes refer to it — existing products and evolve them, and also using those frameworks to create new products,

which would potentially deliver better results, starting from scratch and using these frameworks for accessibility and for Consumer Duty. I do think on the accessibility side of things, just conforming to a framework like WCAG 2.2 or something is only really a checkbox. I think you still need to be testing these products with people with access needs to understand how they're using them, and whether

the way that they set up their screen readers can be used on your website and isn't just a checkbox in a list. So I think human validation, user testing, is still going to throw up loads of really important feedback on these products, as well as just conforming to some of these rules — which are going to be a lot easier to train AI to work through and build products to that specification.

Russell (37:57.039): I did also have a final point. In terms of the job market, I know there's a lot of talk about maybe consolidating roles, and it being difficult for juniors to get into companies and get a starting point — whether there was anything to be discussed around that side of it. Because, if I was a graduate now, I'd definitely be concerned that

what I would have had as the initial work that I would have done somewhere — is that all now being done by the rest of the team using AI? How do I get started in a company where all this junior work is perhaps not required anymore? I do think on that topic, the really talented designers will find a way into the job market. I think you need to be

in that top 20% bracket — they'll always find a way in, and they'll always adapt. And I think the key thing for any young designers is,

approach it with a bit of a blank slate. Because these processes are changing so quickly every day, in some ways it might be easier for them — they join a business and they've got no preconceptions about traditional processes or old tools. They might not be tied to Adobe Suite or Figma, which is kind of difficult perhaps for some designers to move away from. So it's quite an exciting time for those graduates to

just have an open mind. Start building — I think that would be the other thing I'd say. Now is the time for designers to be building and creating more than ever before, using AI. Experiment, build a portfolio, redesign websites, working websites, come up with your own apps, build your portfolio around that — because it's a really exciting time to get some amazing work out there in your portfolio as a young designer.

Mark (40:06.184): Yeah, I think there's always that need for fresh ideas from young people coming into the industry. I guess, in terms of their learning path and understanding tool sets — like you say, it's a different world now. If you've got the ability to sit and prompt and

introduce and interrogate those ideas, you can get something out there quite quickly. You don't need the tool sets to do it. Going back to what we were talking about with graphic designers — there's still an element of training, obviously, to understand colourway and typography and those basic things that you would learn at college. But the ability to just experiment and

challenge the norm with fresh ideas — I think everyone can do, but especially young people, because it's their environment.

I had a really interesting conversation with a couple of guys who work at different agencies last week, actually, in terms of who they're looking to employ. One guy was very much, "We want seniors — we want senior people to come in and slot in, and they can do this job." The other guy was like, "Yeah, but we're trying to encourage juniors in because of this fresh thinking — the ability to,

we can sit them down in front of a prompting program, LLM or whatever, and they can come up with ideas and really challenge the concept of what a senior would effectively think is a straightforward UI that serves a particular purpose. Why not challenge those ideas?" So yeah, I was very much with the young guy

Mark (42:09.98): and that thinking — even though I'm quite old myself.

Amelia (42:14.832): Not at all, not at all. We're kind of coming towards the end of the discussion. So has anyone got any last thoughts? Maybe going back to the original statement, which is, "I'm going to replace my design team with AI." Has anyone got a summary of what we've just discussed in maybe a sentence?

Paul Wood (42:32.739): I'm convinced that we should keep him. Yeah.

Pat (42:33.265): Good luck — it'll probably fail.

Mark (42:37.126): I read a really interesting article today from Don Norman, and he had a comment about it. He said: "Skill set is becoming obsolete; the discipline is ascending — pivot accordingly." That's quite a nice way of putting it, I think. And everyone has to reassess themselves at certain points in their lives, and I think we're at that point, all of us, in terms of what we do, especially within the digital industry.

Russell (43:04.152): I'm obviously incredibly biased on this one, but I think we should be doing more design — but we should definitely be open to doing design in a different way. And yeah, that's probably all I have to say on it.

Amelia (43:24.693): Right then, this is the most exciting bit for me. We are moving on now to jargon busters — this is our weekly feature. Beth and Mark, let me explain it to you. We've got a list of industry terms that you may or may not be familiar with, and I'm going to pick one at random and see if you guys know what it means. Last week, we did all right, didn't we? It was "unicorn" last week, and you were all kind of on the right lines, weren't you?

Paul Wood (43:53.477): Yeah, close to the horn.

Amelia (43:54.958): The horse with the horn, exactly. So this week's phrase is "pump and dump".

Amelia (44:05.217): Does that mean anything to any of you?

Russell (44:07.366): I'm going to go first, because I think it's when there's a kind of meme stock in investing, and you put loads of money into it and you get out before it plummets and you lose all your money.

Amelia (44:21.302): Anyone else?

Pat (44:23.516): It's crypto, isn't it? The crypto bridge, sort of — it happened famously with OpenClaw recently. When the dude who invented OpenClaw, when he first launched it, he launched it under the name ClaudeBot. And then Anthropic issued a copyright injunction against him. So he changed the name from ClaudeBot to MockBot, and then someone jumped on the ClaudeBot

Pat (44:53.788): name, pumped a ClaudeBot meme coin, made loads of money out of it. The price spiked and then crashed again, and the people who pumped that ClaudeBot meme coin made a load of money because they pumped and dumped that coin.

Russell (45:15.79): I think it could just be any stock, though, not just crypto. Wasn't it GameStop? Yeah, there's a really good documentary about it. Wasn't it going out of business or something, and then there was this big Reddit community that kind of got behind it and just, like, increased the price of it — crazy increased the price of it. And then everyone tried to get out. I guess that would be

Pat (45:19.068): GameStop, yeah.

Russell (45:45.336): could be a pump and dump as well — there might be a bit more to that one.

Paul Wood (45:49.188): Hmm, yeah, so it sounds like slang, doesn't it? My best guess would be along the lines of what Pat and Russ have just said — sort of, someone who's actively pushing something with a view to getting out when it's at its peak, I suppose. But it also sounds like a brag kind of term, doesn't it? I can imagine people slapping each other on the back and saying, "That was a good pump and dump."

Amelia (46:16.301): Mark, Beth, have you got any other thoughts?

Mark (46:16.702): Sounds like a city term. Yeah, I wonder, it sounds like a city term, doesn't it? Stocks and shares, that kind of... "We've got a pump and dump today."

Amelia (46:28.535): Something to talk about in industry, maybe.

Beth (46:30.961): Yeah, it's got Wolf of Wall Street vibes for sure.

Amelia (46:34.133): Exactly. So, officially, the definition is: artificially inflating a stock price through false claims, to sell at a profit.

Paul Wood (46:45.073): So it's actually more on the illegal side of things than maybe we were giving it credit for.

Amelia (46:50.517): Yeah, maybe. Yeah, exactly. But we'll have more jargon busters next week, of course. Until next week — our next episode, we're looking at a controversial patent approval for Google. What does that outline, Paul?

Paul Wood (46:52.359): Or at least immoral.

Pat (46:53.316): Yeah, yeah, pretty unethical, yeah.

Paul Wood (47:08.839): So last week I joked about needing an emergency episode of the podcast, and this is probably the first one. Google has been granted a patent. What they've described — and we'll get into it next time — is that, if they deem a landing page on a website to be not up to scratch, they could use AI to create their own version of it, so that the end user sees what they've created and not what the

website owners have created. Now, it's not actually happening right now — it's just a concept that they've had approved — but I think we spoke about it earlier this week in our team and we're all kind of a bit shocked by it. Yeah, it could be quite a juicy topic to get into.

Pat (47:57.436): There are all sorts of compliance headaches, that kind of thing.

Amelia (48:03.593): Looking forward to seeing you.

Russell (48:03.606): Yeah, and branding as well. Branding too.

Amelia (48:07.693): Looking forward to chatting more about that next week. But thank you so much, everyone, for joining us. Thank you for listening, and as I say, we'll be back next week.

Pat (48:18.524): Cheers, everyone. See you.

Russell (48:19.648): Cheers, bye.

Mark (48:21.106): Thanks very much.