Artificial Intelligence in Retail

Scientists across Amazon are revolutionizing how we verify AI systems with automated reasoning. Our teams are making AI more trustworthy, applying the same mathematical logic we’ve used for years to verify security at Amazon Web Services (AWS). It’s called “neurosymbolic AI”, and you can see this in action with robots like the one below navigating Amazon warehouses more efficiently, Rufus providing shoppers with more accurate answers, and AWS teams using this approach for everything from verifying AI-generated code correctness before shipping new Kiro features publicly, to Graviton chip optimization. Last week’s launch of Automated Reasoning checks in Amazon Bedrock Guardrails—catching AI errors with up to 99% accuracy—is another proof point in how we’re making AI safer for everyone. Learn more in the The Wall Street Journal about this game-changing verification technology. https://lnkd.in/g9H-hXnS

Excited to share insights from Walmart 's groundbreaking semantic search system that revolutionizes e-commerce product discovery! The team at Walmart Global Technology(the team that I am a part of 😬) has developed a hybrid retrieval system that combines traditional inverted index search with neural embedding-based search to tackle the challenging problem of tail queries in e-commerce. Key Technical Highlights: • The system uses a two-tower BERT architecture where one tower processes queries and another processes product information, generating dense vector representations for semantic matching. • Product information is enriched by combining titles with key attributes like category, brand, color, and gender using special prefix tokens to help the model distinguish different attribute types. • The neural model leverages DistilBERT with 6 layers and projects the 768-dimensional embeddings down to 256 dimensions using a linear layer, achieving optimal performance while reducing storage and computation costs. • To improve model training, they implemented innovative negative sampling techniques combining product category matching and token overlap filtering to identify challenging negative examples. Production Implementation Details: • The system uses a managed ANN (Approximate Nearest Neighbor) service to enable fast retrieval, achieving 99% recall@20 with just 13ms latency. • Query embeddings are cached with preset TTL (Time-To-Live) to reduce latency and costs in production. • The model is exported to ONNX format and served in Java, with custom optimizations like fixed input shapes and GPU acceleration using NVIDIA T4 processors. Results: The system showed significant improvements in both offline metrics and live experiments, with: - +2.84% improvement in NDCG@10 for human evaluation - +0.54% lift in Add-to-Cart rates in live A/B testing This is a fantastic example of how modern NLP techniques can be successfully deployed at scale to solve real-world e-commerce challenges!

The teen models in Mango's latest campaign have perfect poses, perfect lighting, and one small detail: they don't exist. This Spanish fashion giant launched their Sunset Dream collection using entirely AI-generated models across 95 markets. Not a single human model was photographed. Here's how they did it: 📌 Took photos of real clothes on display stands 📌 Fed these pictures to their AI system 📌 Created model images in minutes 📌 Rolled out everywhere at once The business impact is massive. Fashion brands typically save 60-80% by leveraging AI photoshoots. Those savings can now fund innovation, better pricing, or faster expansion. But cost isn't the real story here. Speed is. While competitors wait weeks for campaign photos, MANGO creates, tests, and launches collections in days. No weather delays. No scheduling conflicts. No reshoots. This wasn't luck. Since 2018, Mango has built 15 different AI platforms across their business. They've been preparing for this moment. The result? Their 2024 turnover reached 3.3 billion euros in 2024, growing 7.6% from 2023. What makes this significant is that Mango proved AI-generated content can drive real sales. Their teen customers embraced these virtual models without hesitation. Fashion's biggest players are watching. If Mango's approach succeeds long-term, traditional photography could become a thing of the past for e-commerce. The brands that adapt now will set industry standards. Those that don't might find themselves competing against companies moving at AI speed. Which fashion tradition do you think AI will disrupt next?

Anthropic just launched something that could quietly change how online shopping works. In partnership with Visa, Mastercard and Accenture, the company behind Claude has released a blueprint that lets retailers build their own AI commerce agents. There are two types of agents: 1. Shopping agents (for customers) These can search a retailer’s catalog, remember preferences, build the cart, and answer questions ....basically acting like a personal shopping assistant inside the AI chat. 2. Merchant agents (for the business) These help store operators by suggesting pricing and promotions, tracking inventory, and analysing sales performance. Shopify and Priceline have already started running live agents using this technology. What makes this interesting is the involvement of the big payment networks. Visa and Mastercard aren’t just watching from the sidelines ...they’re helping shape how AI agents will sit inside the actual commerce and payment flow. That matters, because trust and control are still the biggest concerns for merchants when AI starts interacting directly with customers. Accenture’s research adds another layer: 85% of people say they’re open to collaborating with an AI agent, and nearly three in four would trust a personal AI agent more than their best friend to make a purchase on their behalf. We’re moving from AI that just answers questions to AI that can actually complete parts of the shopping journey. The retailers who figure out how to use this well , while keeping the customer experience transparent ...will have a real advantage. Curious to see how fast this spreads beyond the early adopters. #AI #Anthropic #Claude #Fintech #Ecommerce #Visa #Mastercard #Accenture #Shopify #AgenticAI #DigitalCommerce #Payments

“Boss, we unloaded 87 sacks.” Reality? 100. In factories, warehouses, and logistics hubs, one of the oldest tricks in the book is simple - misreporting. For manufacturers, small discrepancies add up. Over time, theft, errors, and miscounts quietly bleed revenue. At StorePulse AI, we built an AI-powered Loading & Unloading Counter that does what humans can’t - track every single movement, in real-time, without bias, fatigue, or error. ✅ Prevents theft by ensuring what’s reported is what’s actually moved ✅ Eliminates disputes between vendors, suppliers, and logistics teams ✅ Brings transparency to every shipment, every unit, every time No more guesswork. No more blind trust. Just pure data. What’s another manufacturing problem you think AI should solve next?

The traditional Afghan method of preserving grapes in mud-straw containers, known as Kanjna, is a testament to human ingenuity and a deep understanding of natural preservation techniques. While this method has proven effective for centuries, modern technology offers exciting opportunities to further enhance food preservation. What do you think? AI and Technology Solutions: Optimized Storage Conditions: - Sensor Networks: AI-powered sensor networks can monitor temperature, humidity, and gas levels within storage facilities, ensuring optimal conditions for long-term preservation.  - Predictive Analytics: By analyzing historical data and real-time sensor readings, AI can predict potential issues like spoilage or pest infestations, allowing for proactive measures. Advanced Packaging Techniques: - Smart Packaging: Intelligent packaging materials can monitor the condition of the stored grapes, alerting authorities if any issues arise. - Modified Atmosphere Packaging: This technique involves modifying the gas composition within the packaging to slow down spoilage. AI can optimize the gas mixture for different types of produce. Food Safety and Quality Monitoring: - Computer Vision: AI-powered vision systems can inspect grapes for defects, ensuring only high-quality produce is stored.  - Food Safety Testing: Rapid, automated testing methods can detect contaminants and pathogens, preventing foodborne illnesses. Supply Chain Optimization: - Blockchain Technology: Blockchain can track the journey of grapes from farm to table, ensuring transparency and traceability.  - AI-Powered Logistics: AI can optimize transportation routes and storage conditions, reducing spoilage and waste. By combining traditional knowledge with cutting-edge technology, it's possible to preserve food for longer periods, reduce food waste, and ensure food safety. Vai @ Saeed Shah #Ai #Technology #Innovation

Walmart built an AI semantic search system processing millions of queries with 99% recall. - When a user searches for a product, the query goes through a Siamese network with pre-trained tokenisers. This architecture allows the model to use the context of the input queries effectively. - Different attributes are concatenated to the query title using a special token. This ensures that the model can distinguish among different product characteristics, like brand or colour, when processing a query. - During training, the model employs a sampled softmax loss function, where both relevant and irrelevant products are considered for each query -- this helps improve the accuracy in distinguishing between different product matches. - The architecture combines multiple embeddings for both queries and products. This lets the system capture the varying meanings of common queries, improving the model's flexibility and interpretation. Link to the paper: https://lnkd.in/grRHswSX And a great article from Portkey explaining how Walmart also uses Semantic Caching during this search process: https://lnkd.in/gMngD5-T #AI #LLMs #RAG

The news just dropped - and it was only a matter of time. After PayPal and Mastercard, Visa is also going big on agentic AI. It’s called Visa Intelligent Commerce (VIC). Here is my take. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗶𝘁? VIC is a trust layer that lets autonomous AI agents - travel bots, voice assistants, smart fridges - find, decide and pay for consumers. Visa converts an ordinary card into an AI-ready token that: • verifies the agent is authorised by the cardholder • enforces spend limits and rules • uses Visa’s real-time risk models to approve or block each transaction. Visa will extend the infrastructure, standards and capabilities present in physical and digital commerce today to AI commerce. Consumers will enable AI agents via AI platforms to use a Visa credential (4.8 billion today) at any accepting merchant location (150 million) for any payment use case. 𝗪𝗵𝘆 𝗱𝗼𝗲𝘀 𝗶𝘁 𝗺𝗮𝗸𝗲 𝘀𝗲𝗻𝘀𝗲? • AI is moving from chat to action. Autonomous agents are forecast to drive $1 trn in spend by 2030; the missing piece is a trusted “buy” button. • Friction kills sales. Up to 70 % of mobile carts are abandoned; an agent that checks out in milliseconds fixes that. • Visa leverages existing infrastructure built over decades (to combat fraud) and redeploys it for agent-driven commerce. 𝗜𝗺𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀 𝘁𝗼 𝘄𝗮𝘁𝗰𝗵 • Consumers: AI agents embedded in devices - from smartwatches to digital assistants - to shop on a consumer's behalf via programmable spending limits, merchant rules, and tokenised payments. • Merchants & platforms: higher conversion and truly personalised storefronts built for “segments of one” (treating each individual customer as a unique segment). • Banks & fintechs: new AI-ready cards with consent tools and dashboards, monetising agent insights. • Developers: rails-as-a-service; expect an explosion of agent-first apps across travel, retail and SMB back-office - no deep compliance or full-stack checkout flows needed. • Policy & privacy: tokenisation, spend limits, and audit trails offer a template regulators may adopt as autonomous commerce scales.    Visa isn’t trying to build the best AI - it’s ensuring any AI can pay safely. By opening its network as the last mile for autonomous agents, Visa positions itself as the invisible switchboard of the next commerce era. If AI becomes the new browser, Visa wants VIC to become its checkout button. Opinions: my own, Video source: Visa 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 𝐭𝐨 𝐦𝐲 𝐧𝐞𝐰𝐬𝐥𝐞𝐭𝐭𝐞𝐫: https://lnkd.in/dkqhnxdg

We’ve firmly entered the era of Generative Engine Optimization (GEO). AI platforms like ChatGPT, Perplexity, and Google AI Overviews aren’t just influencing discovery. They are becoming the storefront. Consumers aren’t bouncing between tabs, reviews, and product pages anymore. Increasingly, they ask one question and trust the answer. Behind the scenes, AI is doing the searching, comparing, and deciding. Our 2026 Tech Trends highlighted this shift. The storefront is disappearing as discovery, comparison, and checkout collapse into a single AI-native interface. The data already reflects the change. AI-driven commerce traffic is exploding. Websites now serve two audiences: humans and AI agents. Brands are starting to realize they are no longer competing for clicks. They are competing to be included in the answer. This is where GEO comes in. GEO isn’t “SEO for AI.” It is a fundamentally different and constantly evolving operating model focused on how brands appear inside AI-generated responses. That means optimizing for visibility across LLMs, share of voice in AI answers, authority signals, and content that machines can actually understand and trust. Early data shows a clear pattern. The companies gaining the most momentum and becoming critical partners as brands fight for AI awareness are native GEO platforms built specifically for how LLMs read the web. The leaders go a step further. They don’t just monitor AI visibility. They actively shape how models interpret and rank information. If AI is becoming the new storefront, GEO is the new shelf space. Look for the brands and e-commerce sites forming strong relationships with GEO platforms to dramatically outpace their peers. Read more about the forces reshaping commerce (and thirteen other trends) in the CB Insights 2026 Tech Trends report (h/t to Isabelle Lowe): https://lnkd.in/gYmvZtkH

The checkout page is dying. AI agents that can find, book, and pay for things are coming. But how will they actually pay? These are the 4 models👇 Each more autonomous than the last. A breakdown of how AI agents will pay for things: Model 1: Human-in-the-loop 🔁 Your AI finds a hotel, but freezes at checkout. - "I found these options... please enter your card details." - The magic breaks at payment - like early screen-scraping fintech. - This is what OpenAI's Operator does today. Model 2: Stored credentials with approval 💳 ✅ - Your AI can access your saved payment methods but needs explicit permission. - "I found a $199 room at the Hilton. Approve payment with card ending in 4242?" - Better UX, but still requires human intervention for every transaction. Model 3: Virtual cards with controls 💳🎮 - Your AI has a payment method with defined limits. - "I've booked your room using a virtual card within your $250 budget." - Stripe is already enabling this flow Model 4: Autonomous wallet 🤖 - Your AI controls a programmable wallet (think stablecoins). - No checkout flow. No card details. Just seamless transactions. - We have a lot to figure out before this is possible (e.g. fraud? wallets?) This is the holy grail - ambient payments controlled by your agents. The B2B opportunity may be even bigger: • Corporate travel booking within policy • Procurement with spending controls (Model 3) • Treasury management across borders (Model 4) Businesses will adopt these models faster than consumers. The most fascinating frontier? Agent-to-agent payments. - Imagine your shopping agent negotiating directly with merchant agents: •Comparing features • Requesting discounts • Finalizing deals This would transform market dynamics but requires much more infrastructure. We're witnessing the collision of two massive trends: The rise of AI agents x The evolution of programmable money The winners will be those who deploy the right model for the right context. What's your take? If you liked this, I wrote the full piece for the newsletter, You can sign up to at fintechbrainfood .com