This topic is such a can of worms, and it’s perhaps a bit of a departure from the subject matter we usually tackle here on the blog. I think it’s really important to clearly state Grail’s stance on AI, though, because businesses both large and small are increasingly using AI products and services in many ways, especially with regard to their customer communications.

Few things send me straight into “cranky old-timer” mode faster than someone trying to sell me on the benefits of AI. All of a sudden, it feels like every digital service we use for our business operations is touting new AI integrations, whether it’s our point of sale system, website hosting platform, or the software I use for creating graphics and signage. I have “customer management system” companies offering to show me how their AI products can increase our sales by mimicking personalized communication with clients and pestering them at algorithmically-determined intervals. I do not want this. I don’t want ANY of it. I don’t want to use it, I don’t want our customers to be exposed to it, and I don’t want other businesses using it to interact with me as a consumer.

Grail’s foundations lie in face-to-face, human-to-human customer service. Even just five years ago, AI wasn’t really on our radar at all. Our person-centered approach to business was primarily a backlash against both the e-commerce explosion in undergarment retailing, a model that is inherently flawed and fails consumers much more often than it serves them, and the limited-option/ body-exclusionary/ “good luck with that” experience of bra shopping at major chain retailers. There has always been a bit of irony in the way Grail operates because we’re breaking the mold precisely by refusing to follow the tides of online shopping and fast fashion. It’s innovation by way of anachronism, doing things the “old fashioned way” because that’s what actually worked! 

AI is no different, just another manifestation of the push to make sales without the need for any genuine human interaction. This would be gross simply on principle, but it also fundamentally fails the consumer. In no particular order, here are some of the ways Grail pledges to keep our customer relationships authentic and personal:

• I, Carmen Champagne, write every word of this blog myself. It takes time… sometimes more than I really have to spare. There’s a reason why I only post about twice per month. It’s incredibly important to me that these entries reflect my own thoughts, expertise, and values! I don’t pre-create outlines, or do anything else, with AI tools. I’m quite bothered by the fact that I can’t actually stop WordPress from allowing AI models to pull from my content here, despite the fact that Grail has a paid business account.
• All of our graphics and images, whether in-store or on the website, are created by myself using traditional digital tools. I make the featured images for all of these blog posts, except for the one at the top of this article, which was generated by Kim Delatorre using Midjourney. While thinking about this topic, I wanted to find some examples of the often problematic distortions of female bodies common in generative AI art, and I don’t just mean AI’s well-known issues with reproducing human hands. Unattainable notions of ideal body standards have been influenced by edited images since long before social media, but now we’re increasingly being exposed to beauty ideals that are entire fabrications. And yes, I believe that AI art is theft.
• If you ever get a text from me or the Grail team regarding a special order or product request, you are receiving a text from me or the Grail team. The actual people. We send these messages ourselves and receive the responses personally.
• To those companies offering to build us an AI chatbot for text or social media auto-messaging: Go. Away. If someone has a question, they deserve a bone fide response from a human being. It’s not that complicated. How many people actually like it when they have a problem and “customer support” is just a series of recordings or pre-fabricated text responses that do not answer your question or help in any way? Literally no one.
• Our point-of-sale system’s AI tools want to help us streamline our inventory by suggesting “dead stock” to clear out. Except, here’s the thing about being a body-inclusive business… We don’t sell all of our products in all sizes at equal intervals. And that’s okay. We still want to have those sizes and styles in stock for when we do see these customers. It’s part of our foundational values as a business, and it’s how we serve our client community. No thank you; we’re not interested in eliminating our least-sold sizes.
• I’ve been following developments in AI technology for bra-fitting and clothing sizing for several years now. If you’re not aware, there are a number of AI attempts at fitting tools already deployed on the market, both as mobile device apps and in-store with certain large retailers. Of course, AI bra fitting is an incredibly enticing prospect for online retailers and major corporate chains for whom returns are a stubborn problem (up to 40-50% for bras purchased online, depending on the statistic.) Despite many of these tools launching during the Covid era, however, they’ve yet to really make a ripple. Reviews are extremely mixed on their effectiveness, with more petite and smaller-busted users generally having more accurate results. For my part, I’m not at all worried about AI making in-person bra fitting obsolete. The extreme amount of variation in both body shapes and bra shapes, combined with the lack of consistent universal size standards for bras, means that AI will figure out how to draw hands long before it can reliably fit a bra.