The Edit is our weekly show, where Social Editor Grace Robinson quizzes editor-in-chief Ben Hanson on some of the most significant fashion and technology stories from the past seven days.
This week, Grace and Ben discuss AI-assisted holiday hiring and rising delivery costs, regional attitudes to AI, generative imagery’s impact on human modelling, fashion’s bargain with social platforms, and eBay’s results and the growth of secondhand fashion.
This show is available in video format, embedded above, as well as audio, on Apple Podcasts and Spotify.
For deeper interviews with brand and technology leaders, our prestige interview show releases on Thursdays everywhere you find podcasts – and will be coming to YouTube soon.
Subscribers to our mailing list will receive notifications for both shows.
Grace Robinson: Welcome to The Edit from The Interline, the show where we run a quickfire analysis on our pick of the most important fashion and beauty technology stories from the last seven days. I’m Grace, the Social Editor, and I’m joined by Ben, the Editor-in-Chief. Together, we have around 25 minutes to give you our analysis on the stories that we think really matter.
Ben Hanson: Hello, Grace. How are you doing?
Grace Robinson: I’m doing really well. How are you?
Ben Hanson: Still tired. I’m not going to keep inundating the audience with how tired I am week over week, but my advice to you is if you’re thinking about having three children, don’t have three children. I am excited, though, to dig into a mix of stories this week. I think we’ve got some AI in here. We’ve got some holiday shopping in here. We’ve got a mix of different topics, so let’s do it.
The rising cost of delivering the holiday season
Grace Robinson: As we move into the final months of the year, the holiday season is really top of mind for brands and retailers. This week, there’s been a couple of interesting stories that reveal what is really going on behind the scenes as things ramp up.
Firstly, Gap has announced that they’ve launched a new internal AI called Bridgette, and this is essentially made to help support retail hires at the early stages. Since June, nearly 45,000 job candidates have engaged with Bridgette. Essentially, it’s meant to make the hiring process more frictionless and remove some of those more administrative tasks before people actually meet the real human hirer. As Gap prepares for the holiday season, the company has said that this technology will make one of retail’s busiest hiring moments faster, simpler and more human where it counts.
Linking to this, we also saw a report from BoF explaining that we are set to end the most expensive parcel shipping year on record with the highest-ever holiday season delivery prices. This is as delivery services like UPS, FedEx and others are ramping up their surcharges for brands and retailers.
Essentially, the surcharges are meant to reflect the increasing fuel costs, but some delivery services are actually increasing their rates even when fuel costs haven’t risen at all. UPS actually said that the higher price reflects premium services and infrastructure investments that they’re offering.
Now, because of this, some brands are actually switching to smaller delivery companies like OnTrac or GoFo, or encouraging their customers to return parcels in store rather than online. There’s a lot going on here, but what do you think retailers need to think about as we do enter the holiday season?
Ben Hanson: I think you and I will be picking up a couple of holiday stories over the next two weeks or so as we think more about actual markdown strategies and sales and positioning and assortment and things like that. These are both what I’d class as back-office ramp-up technology stories. They’re both really interesting in their own right, and they are both things that brands and retailers do need to have top of mind.
Now, they’re also both tied to cost and cost of living. Both of these things have their roots in the same place. You mentioned that delivery services have increased their prices ostensibly as a result of rising input costs for them, but also as they pursue innovation. That’s directly down to the fact that everything is incredibly expensive, particularly petrol and diesel at the moment.
Then the hiring story: yes, the bit that you see, the bit that you get in the Gap announcement is that they have an AI tool that is making it easier for their HR teams and ostensibly easier for candidates. What they bury the lead a little bit on is the sheer scale of holiday recruitment that is going on at the moment. There’s a statistic in that release where they mention that across all Gap Inc. stores, they receive an estimated 200,000 seasonal employment applications.
That’s a lot. That is a lot of people looking for work. And it’s a lot of people looking for work in a job market that, if you look at the headline statistics, is not terrible, at least at the moment. This seasonal employment stuff is quite often additional work for people. Again, it has its roots in the same place. It has its roots in the cost of living.
There’s a dystopian cynic’s take on this, which is that the economy has become so depressing and the volume of applications that brands receive for this kind of stuff is so high that it’s demoralising for HR and for hiring professionals to deal with this volume of stuff. The other is just the procedural and step-by-step logistical process of it, which is a very valid application of AI.
It does also put you in mind of, I’m sure you’ve seen the stories on LinkedIn, in particular, and YouTube, where people make funny, gloomy videos of being interviewed by AI for roles. Doing 200,000 of those is a bit of a testament to society in some way. I think the statistic that they mention is they scheduled 7,000 interviews out of this total volume. Something like 45,000 people have interacted with it so far; 200,000 are expected to interact with it over time. The volume of people that get through to interview is very small, so there’s that part of it.
Then there’s just the sheer cost to retailers and brands of delivering on the peak holiday season. We look at it, I think perhaps rightly, as the time of year when a lot of volume gets done. There are brands and retailers that structure their entire calendar year around the idea that they might shift 30 to 40% of their annual volume, gross merchandise value, in the November to December compressed window. The opportunity window for the holidays is still significant and is massive depending on the kind of company you are, but the cost of taking advantage of that opportunity is also rising in a pretty dramatic way.
When it comes to the companies that are saying, okay, we will choose not to use UPS or DPD or FedEx or what have you, and will instead go with local couriers and encourage people to return things to stores, I’d encourage people to go back and listen to an interview I did earlier this year with a lady called Oliwia from Decathlon in Poland. Even if you’re not selling in Europe, there’s some fascinating lessons in that one about what it takes to orchestrate logistics, but also the value of meeting people where they are in terms of deliveries and returns and everything else.
I am constantly in awe of what is involved from a technology and a personnel point of view with delivering on the holidays, and these two stories give us a bit of a window into just what exactly is going on behind the scenes.
Why AI feels like an opportunity in emerging markets
Grace Robinson: Next, we have to talk about the differing attitudes around AI from emerging compared to emerged markets. A great survey from GlobeScan conducted for The Economist found that outside rich countries, people are a lot more hopeful and optimistic about AI. Whereas inside richer countries like America or Canada, for instance, people feel a lot more negative about AI’s impact on the future of humanity.
Now, as The Economist did point out in this piece, the reason for this is quite simple. Individuals in richer countries have more access to formal jobs and institutions, so the threat of AI taking that away feels a lot more pronounced. Whereas in poorer countries, AI represents an opportunity for individuals to really help themselves, as they have less access to education, formal jobs and institutions.
An interesting point from this piece that really stood out to me was that small business owners in places like India, for instance, are benefiting from the ability to generate campaigns or ads with AI. That’s another reason why they’re a bit more positive about it.
I know that there’s a lot of nuance to this one, but I still wanted to get your take. Do you think that this optimism will eventually die down everywhere when it comes to AI, or do you think that the West should actually adopt a bit more of this positive attitude towards it?
Ben Hanson: I think it’s hard to tell any entire society to just buckle up and get on with things and quit complaining. I certainly don’t feel that way about AI. I think there’s a lot of very valid concerns about what it should touch, what it should automate, how much of the job market we’re willing to devolve, and so on. You and I have talked about this a ton of times this year. I’ve done a ton of one-to-one interviews with brand and technology CEOs and executives around all of this as well.
I think that the sentiment in markets like Europe, America, Australia, Canada and so on is valid because these are markets that have been knowledge-work-led and service-led for quite a long time. The UK, for instance, is not a regionalised manufacturing hub the way that it once was. It is a service economy, which means that landing a good job in these markets is a matter of doing knowledge work and acquiring the knowledge and the institutional education to get the piece of paper that says that you are permitted and you are sufficiently smart and knowledgeable to go and actually execute on that knowledge work.
A lot of knowledge work, though, doesn’t really rely on that much knowledge. It’s quite procedural and admin-heavy and everything else. I think that’s why the threat of automation and the threat of AI feels so palpable in markets like Europe, the UK, the US and so on. Knowledge work is quite a precarious thing to begin with, and to go after it threatens a lot about the assumptions of those whole economies.
As you mentioned, other markets haven’t had the same access to these kinds of service economies. A lot of them are not set up the same way. People have not had access to the same educational institutions and so on. What feels like a threat in the UK, the US, Europe and so on, I understand why it feels like an opportunity in other places. It’s one thing in America or here in the UK to say, okay, you can run what would be a three-person company with one person with the help of AI, or you can run what would be a five-person company with three people, or a ten-person company with five people. The maths scales.
It’s one thing to say that in markets where people have always had access to these kinds of things anyway, where they’ve always had access to these opportunities. It’s another to open that door in markets where those opportunities haven’t been afforded to people. If you couldn’t run it as a one-person company in Vietnam, for example, you couldn’t run it. Whereas being given access to tools for automation, tools for creative generation, all of these kinds of things, the things that feel risky in developed markets feel less risky over there.
Nobody’s going to change their mind about AI at this stage. If you think about the data that we collected from the survey we did in the spring and summer of this year, people feel the way that they feel about it. I think slow uptake and adoption will erode some of those feelings, but they’re not going to reverse them.
I think in other markets, people see this the same way that they saw access to the internet, which is an opportunity to leapfrog. Entire communities in Africa and Vietnam and places like that had never had good landline internet access and suddenly got mobile internet access, for instance, and it felt like a leapfrog moment for those places.
I think this will manifest itself in fashion in particular, not just because these are manufacturing regions, which they are, but because these are regions that have sufficient manufacturing expertise for people to launch their own brands without the massive overheads that they require to do it otherwise.
I think it’s an interesting story, this one. I would encourage people to go read it if they can access The Economist, because I’m certainly not saying anybody’s wrong to be sceptical or fearful or downright object to AI in the kind of markets that we live in. I think it’s worth bearing in mind the opportunities that it’s opening up to others.
The doors AI is closing for human models
Grace Robinson: Now, on the topic of AI, next we’re going to be talking about AI in modelling. AI in modelling has been a big conversation for the past few years, but it still feels like there is so much uncertainty in this space.
This week, BoF put out a great piece on this, and they revealed findings from Reuters, who actually interviewed 15 model agency directors and reviewed dozens of confidential modelling contracts to see what’s really going on. They found that for one shoot, Diesel actually asked models to agree to alteration or reconstruction of facial features and body proportions. On another contract from Calvin Klein, they found that models were asked to accept, for no extra pay, that the brand may use reasonably unrecognisable images of models to train and fine-tune AI models.
On top of this, the piece also explains how brands like Mango or Zalando are already fully embracing AI in their marketing content, meaning that a lot of models and agencies will lose a lot of revenue as AI models are more adopted.
Now, the British Fashion Model Agents Association has drawn up an AI code of practice, but there is still little regulation in this space, and it’s still really uncertain what the future holds for models and agencies. I wanted to get your take on this. Do you think in the future everything will be AI models rather than real models?
Ben Hanson: Well, it reminds me of two things. One, it’s really hard to talk about human modelling without getting it mixed up with AI models, so the actual wording here is a bit complicated. The other is that this is the exact inverse of what we’ve just talked about. We’ve just talked about AI creating opportunity. This is an overdue and inevitable and deeply unpleasant example of AI doing the opposite, of closing doors.
Now, when I’ve interviewed brand executives, the creative director for Lounge by Zalando was on the podcast not terribly long ago. Very recently, I’ve had a couple of CEOs from generative visualisation companies, companies that specialise in using AI to replace product photography. You get either straight answers or sometimes you get some equivocation from the companies that are doing this.
In the case of Zalando, they have a very good and very clear AI code of practice. In conversation with them, they’ve always been very candid about the fact that there’s still a similar amount of work overall, and time overall, and cost overall in creating an AI-generated campaign. But where that work goes and where that money goes, it flows towards different roles than it has flowed towards in the past. By which we mean it’s flowing away from photographers, it’s flowing away from stylists and makeup artists, and it’s flowing away from human models, people whose job it is to wear clothes, embody clothes and appear in front of cameras on locations.
When you talk to the technology companies, they frame it a little bit differently. I’m generalising here, but broadly speaking, they will say, yes, traditional studio, eCommerce, PDP page photography might be going away as a job for photographers and stylists and makeup artists and models, but some brands at least will do different stuff. They will take the talent that they have, and they’ll go and shoot things they didn’t have the budget to shoot before because they’ve reassigned it from somewhere else.
There’s some validity to that, but I think overall, unquestionably, we are seeing way more generative imagery. We are going to continue to see way more. The data that we collected in the AI Report this year tells us that the vast majority of brands believe that more of their content will be AI-generated or AI-enhanced in 2027 compared to what it is today. It’s inevitable.
Now, as an industry body representing human models, you can enforce contracts. You can encourage people to put provisions into the agreements and the releases that they sign with brands. You can’t force brands to change the language and the verbiage that they put in new releases. I would be very surprised if any brand or retailer that currently has an AI pilot going for generative imagery to put in front of consumers, I’d be very surprised if any of their releases don’t include these kinds of riders now to allow them to go and take existing photography and use it as training and use it as reference material.
That doesn’t necessarily mean that every brand or every retailer is just treating it as a complete free-for-all, and they can take people’s faces and likenesses and poses and what have you and remix them however they want with no compensation. But they do still want to take anonymised versions of those things and use them.
I’ve said this before, but the modelling profession was getting more inclusive. I don’t have any statistics to bear that up, but having spoken to model agencies and things over time, progress was being made in the sense that there were more opportunities for a more diverse set of people to model clothing, either in runway shows, in studio photography, in campaign photography and so on. Those doors are getting closed to everybody now.
I don’t think we’ll have a world where everything is generated. Just last week, we published a podcast interview I did with the brand director for the men’s and womenswear brand TOAST, and they’ve said, you know, we don’t want to use any generative imagery because we think it’s in conflict with the brand. I think that will become a minority position sometime in the next couple of years. What that does to the modelling industry, Reuters and BoF have done the research, and I think it only gets worse from here.
Grace Robinson: Next, we have to talk about the Crafted for Luxury event that Snapchat hosted this week in Paris. The event brought together leaders in luxury as well as Snap content creators. The aim was to show how digital experiences with AR and AI are actually a key strategy to engage younger consumers with luxury brands. At the event, luxury brands like Miu Miu, Gucci and Dior all piloted their own version of a digital experience.
Now, though brands speaking about engaging younger consumers with online or digital experiences does feel like a bit of a broken record at this point, looking beyond Snap, there is actually no doubt that generating brand engagement through digital channels is very much snowballing in the fashion space.
As BoF outlined this week, streamers like Kai Cenat have really blown up in the fashion industry. Brands such as 424, Burberry and H&M have all invited streamers to sit front row at their runways and even walk their runways. Even sportswear brands like Adidas and Under Armour have recently partnered with streamers on sneaker collaborations.
The piece also highlights how livestreaming viewership crossed 36 billion hours in 2025, and it’s only set to grow more. There’s no doubt that these kinds of digital channels are a massive opportunity for fashion brands.
But on the flip side to this, we also have to address the fact that there’s been a lot of conversation around how any content shared on Meta platforms can actually be used to train Meta’s products. Again, there’s a lot of nuance with this story. What do you make of it, and what do you think fashion brands need to be thinking about?
Ben Hanson: I’m going to give you a really quick answer because we’re going to go way over this week’s time, I think, if I don’t. There’s a complicated relationship, I think, between social platforms and brands. The stuff that you mentioned where you have streamers from Twitch and Discord and places like that launching their own brands, great. It makes sense where people have an audience, they then see an opportunity to turn that audience into products.
Snapchat, not exactly the social platform with the biggest impact of all of them, and certainly not the best run of all of the businesses here, does have an enduring reputation. It does have an enduring audience, and I think it does make it a logical target for acting as an intermediary between brands and consumers.
The thing with all of this is every one of these is a closed platform. Every one of these will say, oh, we will help connect you to a wider audience. We will give you more distribution and more exposure than you would have had otherwise. That distribution and that exposure is controlled by them. The terms and the conditions are controlled by them. Brands go into this bargain understanding that.
But the Meta AI training, which businesses apparently have a way to opt out of, at least in the short term, is a reminder of the fact that unless you own a distribution channel, or an inspiration channel, or a product creation channel, or anything else, if it’s in the hands of somebody else and it’s algorithmically led, you are at the whims of whoever is the owner of that platform.
I think brands are going to continue to look at these as sources of inspiration, ways to find new talent, and opportunities to create things that the market actually wants. The knife will twist over time, and you will be reminded of the fact that all of the work that you’ve done there belongs to somebody else, and the control of it belongs to somebody else over time.
I don’t know how the whole social network backlash or anything is going to play out over time. I think it ends up with more balkanised, interest-based networks, to be fair, which Twitch is and which Discord is, rather than the big social ones. I don’t personally use social media outside of work. I have limited exposure to this. I do know who some of these streamers are, so I’m going to take that as evidence of the fact that I’m not maybe as old as I thought I was.
Can eBay win the secondhand fashion market?
Grace Robinson: Finally, let’s talk about eBay’s third-quarter results. Revenue is up 14% to $3.2 billion, so the business is really going from strength to strength. It’s fair to say that the success is really highlighting the growth of the secondhand fashion market.
For example, in the brand’s sixth annual Recommerce Report, it found that 47% of consumers surveyed purchase secondhand goods at least monthly, and that’s up from 35% in 2025.
Additionally, Depop is also doing really well since eBay acquired it. eBay’s livestreaming is doing really well, and the brand has also just expanded its Authenticity Guarantee services for luxury fashion across Europe. Again, things do seem to be getting better and better for eBay. Talk to me about this landscape and also how you think technology is really driving growth in the secondhand fashion market.
Ben Hanson: This one’s an interesting one because you have specialised fashion companies that do this. Vestiaire Collective has its own niche. It is very specifically about the resale of premium and luxury fashion. Then you have StockX, which at least to start with was very focused on streetwear and scarce hype drops and sneakers and things along those lines. They’re all terrified of eBay.
This is the thing that comes up whenever I have a conversation. I just recorded one yesterday for the weekly interview show, which will be out in a couple of weeks, which has a resale component to it. It’s not the main thing. As part of that conversation, it came up that there was a market opportunity for the platform in question to do its own resale. The same opportunity that every brand has, the same opportunity every retailer has. You can do this yourself if you want to. You can start a secondary market. The answer that came around was, well, sure, that’s interesting. We have a really good partnership with eBay.
Of course you do, because eBay is synonymous with the secondhand market and has been for a very, very long time. I know Vinted has good scale here as well, but it’s eBay’s game to lose, I would say, in the long term. I think they’ll continue to have the specialist Vestiaire Collectives and so on, but the amount of money that’s flowing into resale, the amount of attention that’s flowing into it, I think eBay are well positioned to do it. Not just because of the Depop acquisition, but just because the eBay name is synonymous with a lot of this.
Interestingly, though, they made a lot of money in the quarter in question, but they actually also spent the vast majority of it on sales and marketing and AI initiatives and things for Depop. Depop in particular has become a cost centre for them as well. That’s not because it was a bad business when they acquired it. That’s because they see a massive opportunity to continue to scale fashion resale. They’re in the position to do it.
You and I talked about this last week. The financial incentive is what’s driving so much of this resale attention now. It’s the fact that people can generate revenue from clothes that they don’t need anymore. It’s the fact that reasonable-quality, premium and upwards products can appreciate in value if they’re properly maintained and repaired. I can’t remember the exact statistic from last week, but the cost of repairing things is beneficial over time because you can then resell things that have been well repaired and well maintained for more afterwards.
There’s a big shift here that’s happening with the technology platform because of eBay, because of Depop, because of the investment in AI and everything else. But to bring us full circle, it’s happening because of the cost of living crisis. It’s happening because people can’t acquire things firsthand or don’t want to acquire things firsthand. Maybe this prediction will age badly, but I think I have a sneaking suspicion that eBay ends up winning the secondhand fashion game in the long term.
