AI Agents & the Future of Personal Shopping - Part Ⅱ

Shopping just crossed a line: the bot doesn’t only recommend, it remembers, compares, and buys with your permission. We unpack a pivotal year where AI agents went from reactive chat to proactive concierge, reshaping how intent turns into a purchase across Amazon, Walmart, Google, and OpenAI’s expanding ecosystem.

We kick off with Amazon’s Rufus and its late‑2025 leap: activity-aware recommendations, hands-free price‑drop purchases, cart‑to‑doorstep execution, visual search that converts a messy fridge into a grocery plan, and list autopilot from a scribbled note. With model orchestration, real‑time retrieval, and side‑by‑side comparisons, Rufus aims to reduce doubt and win checkout with transparency and speed. Then we shift to Walmart’s Sparky, a companion that meets you in the app and, through a bold OpenAI partnership, inside ChatGPT. From occasion dressing to meal plans and service bookings, Sparky embodies “agentic commerce,” anticipating needs and acting across channels without friction.

Google’s AI shopping mode brings Gemini-powered results directly into search: natural-language queries, review summaries, live price tracking, visual try‑ons, and even “let Google call” to confirm local inventory. With early tests of AI-led checkout, Google compresses the journey from discovery to payment, keeping trust by emphasizing relevance over ads. Finally, we dive into OpenAI’s instant checkout and the open Agentic Commerce Protocol with Stripe and PayPal, opening conversational buying to Etsy, Shopify merchants, and beyond—bringing small businesses into chat-driven commerce without custom builds.

We spotlight women-led startups redefining the experience: Phia for radical price transparency and secondhand value, Gensmo for mood-driven fashion and real-time avatars, Alta for closet‑aware styling, and Doji for social, hyper‑real try‑ons. Along the way, we map how shoppers adapt—verifying AI advice, embracing seamless purchases, and blending AI groundwork with human nuance—as trust features like citations and comparisons close the last mile to buy. If you care about the future of retail, personalization, and the new rules of intent, this conversation gives you the playbook.

Enjoyed the show? Follow, share with a friend who loves great tools, and leave a quick review—what would you let an AI buy for you next?

Transcript

Setting The Stage For 2025

Speaker 0:04

Hello and welcome back to She Builds With AI, where we explore how women and technology are shaping the future. I'm your host, Yunia Lach, and this is part two of our series AI Agents and the Future of Personal Shopping. In our last episode, we introduced the rise of AI shopping assistants and how nearly every major retailer started deploying bots to streamline how we discover and buy products. That was the year of the AI Assistant. And today we're jumping into late 2025's developments. A lot has happened since mid-2025. From major upgrades by retail giants to a wave of new startups and shifts in shopper behavior. So grab a coffee, settle in, and let's dive into the latest trends. Let

Amazon Rufus Grows Up

Speaker 1:01

us start with Amazon, which has dramatically beefed up its AI shopping assistant Rufus. Launched in early 2024, Rufus was already helping users ask natural questions and get tailored product suggestions. But by late 2025, Amazon reports it's smarter, more capable, and more helpful than ever. In fact, Amazon says it rolled out 50 plus upgrades to Rufus' intelligence and reasoning, making it the next gen agencies for shopping. What does that mean in practice? First, more ways to shop. You can now ask Rufus to find products by activity or occasion. Say, I need gift ideas for a 10-year-old's birthday party, and it will parse that context to recommend age-appropriate toys, games, and attire. It's far beyond keyword search. Rufus understands your purpose. Second, truly hands-free buying. Rufus isn't just a chatty salesperson, it's becoming an autonomous shopper. Prime users can set a price threshold on an item and Rufus will auto-buy it once the price drops to your target. For example, buy this jacket if it goes under a hundred bucks. Rufus will do it, saving you money automatically. It even tracks price histories with 30 and 90 day charts. So you instantly know if a deal is good. Third, cart to doorstep. The assistant can now add items to your cart on command. Check if you're getting the best price and even complete the purchase on your behalf in some cases. In other words, it's become a concierge that doesn't just advise but acts. Fourth, visual shopping and list autopilot. Rufus gained a visual search ability. You can upload a photo like a messy fridge or a torn rug and ask for recommendations, groceries to restock, a cleaner for that stain. And here's a life simplifier. If you scribble a handwritten shopping list, Rufus can scan the photo and automatically add all those items into your Amazon cart.

Hands‑Free Deals And Autopurchase

Speaker 3:16

No more manual entry. Behind the scenes, Amazon supercharged Rufus with a blend of advanced AI models. It taps Amazon's own Nova model plus Anthropics Card and a custom model all orchestrated via AWS Bedrock to choose the best model for each query. It also uses real-time retrivial to pull in up-to-date info from trusted sources like the New York Times or Vogue when you ask broader trend questions. In short, Rufus has a massive knowledge graph of products and a memory of your preferences. Amazon says Rufus now remembers details like your pet's breed or your kids' ages, the personalized suggestions, think gift ideas for a five-year-old who loves soccer. It's like a personal shopper who never forgets anything you've mentioned. And users are responding. As of late 2025, over 250 million customers have tried Rufus with monthly users of up 149% and interactions up 210% year over year. Executives disclosed that Rufus is on track to add $10 billion in annual sales by guiding more purchases. Shoppers who chat with the AI are 60% more likely to buy something in that season. That's huge. Amazon even shared in an earnings call that they've increased their investment in AI infrastructure, like a new 1 billion data center, partly due to Rufus' success. Clearly, they see this as the future of shopping. However, not all feedback is rosy. Some users initially found Rufus could hallucinate, making up facts or parroting product descriptions too directly. Amazon has worked to improve factual accuracy, like the rack approach, pulling real reviews and articles. They even added new help me decide features that gives you side-by-side comparisons on products to build trust. The company acknowledges that winning consumer trust is critical. They don't want roofers giving bad info and scaring shoppers off these tools. For now though, usage keeps climbing, especially during big sales. Shoppers are clearly curious and increasingly comfortable using AI agents to sniff out deals and gift ideas during peak season. Next up, Walmart. The retail giant has been busy crafting its own shopping companion, Sparky, and forging a partnership with OpenAI to meet shoppers wherever they are. If Amazon is keeping customers on Amazon.com, Walmart is saying we'll go to you, even if that's on Chat GPT. Let's unpack this. Sparky launched in June 2025 as Walmart's AI assistant, and it's directly embedded in the Walmart mobile app. Look for the smiling Ask Sparky button. Sparky is build as a true AI-powered retail companion that can do a bit of everything. For example, occasion-based recommendations, planning a

Visual Search And List Autopilot

Speaker 6:32

party or prepping for a trip. You can ask Sparky for ideas. For example, I'm going to a beach wedding. What should I wear? Sparky will not only recommend an outfit, but it can check the beach destination's weather and suggest appropriate attire. It's context aware, pulling in info like sports schedules or climate if needed. Another example review, summarizer. Walmart has tons of products and reviews. Sparky uses Gen AI to synthesize product reviews for you, so you don't have to slog through hundreds of comments. Also, comparisons and QA. You can have a back and forth with Sparky. Like which 55-inch TV is best for a bright room? It will compare options and even follow up with clarifying questions, much like a helpful sales associate. The goal is that you add to card with confidence, feeling you've gotten personalized advice. Soon there will be multimodal and more. Walmart hints that Sparky will soon go multimodal. Understanding images, voice, maybe even video. Also on the roadmap, auto reorders of your staples, booking services, imagine asking Sparky to reschedule a tire installation or carpet cleaning along with product purchases. And generally being proactive in your daily shopping needs. In short, Sparky aims to combine a personal stylist, shopping assistant, and customer service rep all in one. Walmart frames this as part of the future of shopping is agentic, a term they use to describe AI evolving from a reactive chatbot

Models, Memory, And Trust

Speaker 8:18

to a proactive, smart agent. They want Sparky to anticipate needs. For example, you say, what's for dinner? Could trigger a week of meal plans with ingredients auto-added to your cart, or how do I fix a leaky faucet? Could result in step-by-step guidance plus the parts you need scheduled for some day delivery. It's almost like having a handy friend on call who can also do your shopping. Now beyond its own app, Walmart made a bold move in October 2025. It partnered with OpenAI to integrate shopping into ChatGPT. This is huge. It means soon you can link your Walmart account to ChatGPT and literally buy Walmart products from a ChatGPT conversation without ever opening the Walmart site. This was announced as part of OpenAI's new instant checkout program, which we'll discuss more in a bit. Essentially, if you're chatting with ChatGPT and it suggests an item including Walmart's inventory, you can say, yes, buy that for me, and the purchase will go through right within the chart. This identic commerce in action bolts becoming personal shoppers across platforms. For Walmart, the strategy is twofold. Enhance their own channels with Sparky and extend reach to external channels like ChatGPT. Walmart's stock even ticked up on the news of the OpenAI partnership, showing investors recognize the significance. By linking with ChatGPT, Walmart is effectively opening a new front door for shoppers. Walmart coined a great phrase for this proactive AI-powered shopping, agentic commerce. They describe it as AI moving from reactive to proactive, from static to dynamic, learning and predicting customers' needs. Instead of you searching and browsing aimlessly, the agent plans and curates for you. This partnership is one of the first real examples, from inspiration to purchase, all in one conversational flow. It's worth noting Walmart isn't the only one eyeing ChatGPT as a sales channel. We're seeing a trend of retailers hooking into AI chat platforms. Target, for example, built an AI gift finder. And others are experimenting too. But Walmart's move with OpenAI is among the highest profile. It basically says we don't mind if the sale happens on Walmart.com or in a chat GPT window as long as we fulfill it. They're chasing convenience and customer reach, and they're doing it while still improving their own app. Now let's talk Google, the other behemoth in this race. In part one, we mentioned Google's early foray into an AI-powered shopping mode, which promised features like visual try-ons and the buy for me button. By late 2025, Google has fully rolled out these capabilities, positioning itself as a personal shopping agent within its search engine. If you've

Walmart’s Sparky Strategy

Speaker 11:29

used Google recently for shopping queries, you might have noticed a new conversational AI mode available. Just ahead of the holidays, Google announced a batch of new AI shopping tools. Here are the highlights. US users of the Google app, now powered by Google's Gemini AI model, can enter natural language queries and get shoppable results. Google's AI will return a dynamic list of recommendations with product images, specs, real-time prices, and even summaries of reviews. This makes shopping searches feel like asking an expert rather than searching a catalog. And importantly, these results span over 50 billion product listings that Google tracks, updates constantly. Google has its massive shopping graph in the background, so the AI has a deep well of product data to draw on. One very nifty few feature is let Google call. If you're looking at a product that's sold in a local store, Google's AI can call the store for you to check if it's in stock, confirm the price, or even ask about current promotions. Imagine surging for a specific Lego set. Google can ring up the nearest toy store to make sure it's available right now. It's like having a personal assistant who will do the legwork of phoning stores so you don't have to. This kind of agent action blurs the line between online convenience and real-world inventory. Built in price tracking and comparison. Google's AI mode now lets you set price drop alerts with a simple tap see something you like, hit track price, and the AI will ping if the price falls, say during a sale. It will also surface a comparison table, giving you a neat breakdown of features and prices. In short, Google is leveraging its data crunching to be the ultimate comparison shopping assistant automatically. Visual try-ons and style ideas. Earlier in 2025, Google introduced AI-driven virtual try-on for clothing. You can now upload a photo or use Google Lens to see how an item in your size might look on a real model that matches your body type. They've even added a vision search where you can generate a look, for example, show me ideas of outfits for a winter holiday party. And the AI will produce an inspiration board and images and link the pieces so you can shop those looks. It's part of making search more visual and creative, not just textural.

Agentic Commerce Goes Cross‑Platform

Speaker 14:16

Essentially, Google's searching by vibe and style now, not just keywords. Perhaps the biggest development is that Google is now testing its own version of AI-driven checkouts, akin to OpenAI's instant checkout. Google said it began rolling out a GenTic AI checkout with select US retailers in late 2025. Partners like Wayfair, Chewy, and some Shopify merchants are among the first allowing Google's AI to facilitate purchases. This means if you find a product via Google's AI recommendations, you might be able to buy it then and there through Google's interface without being kicked out to the retailer's site. The details are still emerging, but it's clear Google doesn't want to be left out of the instant buy trend that ChatGPT kicked off. They have immense reach, and if they can keep a user's whole shopping journey within Google, from search to payment, that is powerful. Google's emphasis is that all this is unsponsored and user-first results are ranked by relevance and quality, not ads. And multiple factors like availability, price, seller, reputation, etc. are considered. Nearly two-thirds of consumers have used a general AI like ChatGPT or Google's Bard to help with online shopping decisions as of 2025. And over half said their searches have become more conversational in style. That backs up why Google is doing this. People want to search in natural language and get direct, personalized answers. The old way of combing through a dozen websites is waning. Google even noted that as these AI tools spread, the flood of AI-driven options is reshaping the shopping experience fundamentally. It's trainable, personalized, and everywhere you need it, including calling a brick and mortar store on your behalf. So Google is effectively turning search and the shopping tab into a personal shopper that's on call 24-7. With Gemini, their next gen model, powering a lot of this, Google's vision of an AI-assisted shopping search really came to life. And it's interesting, we now see Amazon trying to keep users from needing to go to Google via Rufus deep integration and even citing outside sources within Amazon's answers, while Google is trying to keep users from needing to go to Amazon or retailer sites by handling more of the purchase floor itself. So the race is on to become the go-to AI shopping assistant, whether that's through your search bar, your shopping app, or your voice device. So let's dive deeper into OpenAI's instant checkout. A late 2025 development that has major implications for personal shopping. In September 2025, OpenAI announced it was taking the first step towards agentic commerce in ChatGPT with a new feature called Instant Checkout. This lets people buy products directly within a ChatGPT conversation. Initially they launched it with two big integrations, Etsy

Google’s AI Shopping Mode

Speaker 18:08

and coming soon, Shopify's network of merchants. Here's how it works. If ChatGPT suggests a product to you, for example, I recommend this handmade ceramic mug from Etsy. You'll see a buy button next to the suggestion. Click Buy and ChatGPT will prompt you to confirm your shipping and payment details. It can use your saved payment info if you're a ChatGPT Plus user, for instance. Once confirmed, the order is placed right then and there without you ever visiting Etsy's website. A few taps in the chat and you're done. What's revolutionary is how they're doing this. OpenAI in partnership with Stripe and others developed an open standard called the Agentic Commerce Protocol. It's essentially a set of APIs and definitions that let AI agents like ChatGPT interact with merchants' systems to perform transactions on a user's behalf. They open sourced ACP that any merchant or developer can start integrating their store with AI assistance. OpenAI describes ACP as a language that lets AI agents and business work together to complete a purchase for a user. Think of it as a universal translator between chatbots and a shopping cart. By open sourcing it, OpenAI signaled that they want a whole ecosystem of AI commerce, not just a proprietary feature. The trust and security aspect is huge. When you buy via ChatGPT, you're not handing your credit card to a random bot. The transaction is handed off to the merchant's existing payment processor. For example, OpenAI partnered with Stripe on the back end and later with PayPal to handle the heavy lifting of payment processing. In fact, in October 2025, PayPal announced it will adopt the Identic Commerce Protocol to bring tens of millions of PayPal's merchants into ChatGPT Commerce. PayPal essentially built an ACP compatible gateway so that any small business that takes PayPal cloud potentially have their products show up and be bought in ChatGPT. This is a big deal. It means AI-driven shopping isn't limited to a few big retailers. It could empower lots of small businesses to have a presence in conversational AI interfaces. By late 2025, OpenAI's instant checkout was still in early partner rollout. Etsy in the US, live, Shopify coming, others in application. But it's expanding. It supports only single-item purchases at first, but multi-item card support is on the roadmap. And it's free for users, no markup on prices, no subscription needed. Merchants pay a small fee for transactions, similar to a marketplace commission. The broader agentic commerce concept here is that your AI assistant can truly act on your behalf. We've never really had that before at scale. It's always been search for item you click and buy. Now it's ask AI for advice. AI finds item, AI buys item for you with your approval. It closes the loop. This raises fascinating possibilities. For example, you might give ChatGPT a monthly budget and some preferences and let it autonomously replenish your household goods when you're low. Essentially a personalized AI-driven subscribe and save. OpenAI's CEO Sam Altman has spoken about autonomous agents being the next breakthrough where you can delegate

Try‑Ons, Price Tracking, And Checkout

Speaker 21:49

complex tasks with minimal effort. Instant checkout is a step in that direction. Autonomous shopping agents. From the user perspective, early testers have been excited. Imagine you're chatting find me a unique anniversary gift under $50, and the AI finds a handcrafted item on Etsy that's perfect. It can say, shall I purchase this for you now? And you say yes, done. This convenience is likely to shift user behavior further, removing even the slight friction that remained in online shopping. However, not everyone is embracing this open agentic future. Before we get to the tension that's brewing, we'll talk about Amazon's fight with an AI startup in a moment. Let's look at the startup scene. Late 2025 has seen a surge of new companies reimagining personal shopping with AI. Many of these are founded or co-founded by women and are receiving significant funding. Here are a few standout startups to know, each with a unique approach. Phia, your personal dealer hunter, launched in April 2025 by co-founders Phoebe Gates and Sophia Chiani. Phia is an AI-powered shopping agent focused on saving your time and money. It's like a super smart shopping search engine that compares prices across 40,000 plus retailers, finds coupon codes, checks resale values, and tracks price drops in real time. In just five months, post-launch, Phia amassed 500,000 users and partnered with 5,000 brands, driving tens of millions in sales, a testament to how hungry consumers are for price transparency. They raised an 8 million seed round led by Kleiner Perkins to scale up. Phia's Vision is a new search engine for shopping that learns your preferences and surfaces the best deals for what you want. Notably, it isn't just retail. Phia is one of the largest secondhand product databases, 300 million items indexed from resale marketplaces to help users decide if buying used is a better value. This appeals to the Echo and Budget Conscious shoppers. Going forward, FIA is developing a personalized AI shopping advisor, fine-tuning one millions of user data points, aiming to proactively recommend items you love within your budget, with backing from high profile investors, yes, it's Bill Gates' daughter at the helm, plus support from folks like Hailey Bieber and Sheryl Sandberg. Phia exemplifies the new wave of women-led AI startups tackling commerce. Gensmo, a fashion guru with deep pockets. If Phia is about deals, Gensmo is about style and experience. Founded by Ning Hue, a former Google AI lead and Alibaba VP, Gensmo raised an eye-popping 60 plus million dollars in seat and angel funding by mid-2025. An enormous vote of confidence. Gensmo's app is an AI

OpenAI’s Instant Checkout Explained

Speaker 25:06

native fashion shopping platform that makes finding your personal style easier, smarter, and more fun. One hallmark feature, a real-time virtual try-on using digital avatars. You can see outfits visualized on a 3D avatar of yourself instantly, mixing and matching pieces from any brand. Planning a trip? Gensmo will help you build a capsule wardrobe showing how each outfit looks on your avatar. It's focused on what founder Ning Hyu calls helping people feel good in what they wear rather than just efficiency. Under the hood, Gensmo integrates deep learning, computer vision, and generative image models to create personalized mood boards and outfit recommendations tailored to your taste. It even has a vibe image feature that lets you conjure up a fashion scene, say Parisian Cafe Autumn Look, and it will generate an image and item suggestions matching that aesthetics. Importantly, Gensmo also connects to a broader network of e-commerce partners so you can buy within the app. No more juggling tabs or guesswork on sizing. Gensmo users generated over 10 million outfit matches and browsed 100 million products. And the app has been a hit on TikTok, Instagram with its community sharing looks. The combination of serious AI pedigree. The team includes AI PhDs and fashion industry vets, and social media buzz gives it a ton of momentum. It positions itself as your fashion AI agent for everywhere. Highlighting how these agents can specialize in this case, making you confident in your style. Alta, clueless closet, now a reality. Fans of the news movie Clueless might remember the computer program that helped pick outfits. Alta is basically that dream come true. Founded by Jenny Wong and launched in mid-2025. Alta is an AI-powered personal stylist that works with your own digital wardrobe. They raised 11 million seat dollars, led by Menlo Ventures, with investors like the All No family of LVMH and fashion insiders to build the next generation of personal shopping and styling. Alta's app lets you upload your closet or purchase history and then generates outfits recommendations for your daily life. Considering your lifestyle, upcoming occasions, local weather, and budget. It's not just generic fashion advice. It might tell you pair your black blazer that you bought last year with these new trousers for your meeting tomorrow. And by the way, it's going to rain, so grab those ankle boots. Alta uses dozens of in-house multimodal models trained on fashion, data to understand clothing at a granular level. A standard feature is the virtual dressing room. You can try on outfits virtually by seeing them on avatars of yourself. And even mix your own wardrobe items with new suggestions in one view. For example, see how that new jacket from Zara would look with these jeans you already own. Alta essentially brings a personal stylist

The Agentic Commerce Protocol

Speaker 28:21

experience, once a luxury service, to anyone with a smartphone. It's catching attention, the Council of Fashion Designers of America partnered with Alta to explore how AI can enhance styling at scale. The mission, Ulta says, is to make everyone look and feel their best by decoding fashion taste and giving truly personalized advice. From an investor standpoint, Alta's appeal is that it bridges hardcore tech and consumer-friendly design. One VC noted Jenny Wong's rare blend of deep technical expertise and intuitive grasp of fashion trends. Rounding up our startup tour is Doji, which targets the fun and social side of shopping. Just launched in May 2025, Doji had a viral debut. Days after appearing on the App Store, it raised 14 million seed, led by Five Capital and 776, Reddit co-founder Alexis O'Hanean's firm. Doji is all about making virtual tryons engaging by letting you create a hyper-realistic avatar of yourself and then experiment with outfits on that avatar. The founder Dorian Dargan and Jim Winkins, hence the name Doji, come from ARVR backgrounds. Apple's Vision OS, Google's Deep Mind, etc. They saw how much people loved creating avatars. Remember the bus around lenses. AI portraits and applied that to fashion. Using their own diffusion models, Doji generates an avatar from just a few photos of you with uncanny realism. It's like looking in a digital mirror. Once your avatar is ready, the app serves up looks, full outfit ideas displayed on you, or rather your virtual twin. You can swipe through outfit options as if you're playing paper dolls with yourself. Users have said it's surprisingly inspiring to see yourself in clothes you wouldn't normally wear. Doji also has a social layer. You can share your avatar's looks with friends for feedback, almost like a virtual fitting room. Doji puts the shopper at the center and makes it shareable fun. Technically, Doji's approach is intensive. It asks for 8 photos and takes 30 minutes to generate your model, aiming for accuracy over speed. But the result is a high fidelity avatar that if you don't like how it turned out, say it made you a bit too tall or slim, you can retrain with more images. Doji's philosophy, if trying on clothes becomes as easy as scrolling Instagram and as true to life as looking in a mirror, it will transform e-commerce and even reduce returns since you'll know how it fits and looks. They're in invite-onde, due to demand and computing costs, but the bus is real. Early adopters have been impressed with the tech's realism. Keep an eye out for doji as a glimpse of how social AI can turn shopping into a kind of game

Small Merchants Join The Chat

Speaker 31:39

with you as the main character. So many exciting startups, and notably many are founded by women or have women leading innovation in this personal shopping AI boom. They're tackling everything from cost saving to styling to immersive triads. And investors are pouring money in, seeing that the way people shop is on the cusp of a major tech-driven shift. All these advancements back the question, how are consumers adapting? The latter half of 2025 is giving us the first real indicators of evolving shopping behavior in the age of AI assistance. Here are a few trends and patterns emerging. Conversational shopping is becoming the norm. Surveys show that the majority of shoppers are now comfortably using conversational AI for shopping assistance. Retailers are responding to integrating chat interfaces on their sites and training their AIs to handle these detailed queries. Shoppers trust AI more for recommendations but still verify. The increase in usage of tools like Amazon's roofers shows that shoppers are finding value in AI suggestions. Particularly for product research like comparing complex electronics or finding gifts, many now start with an AI query rather than going straight to Google or Amazon search. However, when it comes to hitting that buy button, some caution remains. Conver Conversion rates via AI assistance still slightly trail traditional e-commerce flows. Though the gap is narrowing, this suggests people often use AI to research, then either click through the purchase or double check the details themselves. As AI outputs become more trustworthy, citing sources clarifying uncertainties, consumers are gradually relinquishing that final bit of hesitation. Features like trusted reviews, summaries, or external citations are intentionally there to build confidence so the user doesn't feel they're blindly following a robot's pick. Seamless purchasing is boosting impulse buys and multitasking. Early evidence from OpenAI's instant checkouts and Walmart ChatGPT integration suggests that frictionless purchasing in chat can increase spontaneous buys. As shoppers see what's possible, virtually trying on products, getting price drops, alerts without effort, their expectations are rising. They'll start to gravitate to platforms that know them and assist proactively. The idea of one size fits all shopping site might feel outdated. We're moving into an era where your version of an app might be totally different from mine, because our AI shopping agents shape it to us. Retailers who do this well, like remembering personal details

Startups Redefine Style And Deals

Speaker 34:32

or customizing to a user's tastes, will build loyalty. Those who don't will start to feel impersonal. It's not all smooth sailing. A segment of consumer voices, concerns about AI advice, can they trust it? Does it understand quality and nuances like fabric feel or true color that a human would? There have been anecdotes of AI assistants pushing the highest priced item or misinterpreting a preference. Retailers have noted this and often provide an easy opt-out to a human. Successful deployments tend to be hybrid. AI handles the groundwork, but a human can step in for complex or sensitive cases. Over time, as AI proves itself, and as younger AI native consumers become the majority, trust should grow. But right now in 2025, we're in the trust building phase. Shopping is a collaborative, interactive experience. Apps like Doji indicate people enjoy making shopping social again, even if virtually. We see users creating and sharing lookbooks or asking AI to generate style options and then pulling friends on which to buy. Also, AI can play stylist in group chats. For example, a group of friends can collectively chat with an AI to plan matching outfits for an event. This interactive, almost playful mode of shopping is emerging, moving away from the solitary scroll and buy. Retailers are experimenting with AI-hosted shopping events. In summary, shoppers are rapidly acquiring a digital sidekick mindset, using AI agents to outsource the boring parts of shopping, research, comparing, checking stock, so they can focus on the fun parts, getting exactly what they need, discovering cool new things, and doing it efficiently. The latter half of 2025 shows that many shoppers are on board. They're testing the waters with these AI agents, and many are already hooked on the convenience and personalization. Coming up in part three, we'll continue this journey by looking at the bigger picture and a few storm clouds on the horizon. We'll discuss agentic commerce more deeply, how open AI and others are setting standards for AI shopping agents and a brewing legal battle between Amazon and an AI startup that could shape the rules of engagement. Plus, we'll hop around the globe to see how international retailers, from China's Alibaba to Europe's fashion brands, are innovating with AI assistance. Don't go anywhere. Part three will wrap up our series with a forward-looking bang. You're listening to She Builds With AI Podcast. I'll be right back with a final part of our special on AI and the future of personal shopping. And the final episode for this year. So stay tuned and keep building and keep innovating.