AI brings huge change to the retail/e-commerce industry
Today, retailers face many of the same challenges they’ve had for years, and retailers have made good progress. By focusing on specific problems including managing inventory/supply chains, labor shortages, and shrinkage, AI will allow them to execute even better. There has been a great deal of focus on customer loyalty and engagement, and we discussed how digital native retailers were setting new standards for digital shopping. Now we are entering the AI era, where we have the technology and the know-how to make real improvements in these areas. Retailers are embracing AI in a big way, giving incumbents a run for their money.
The majority of retailers are increasing their technology spend to improve self-checkout, reduce shrink, decrease perishable loss and keep both shoppers and employees safe—and they’re anticipating AI as a big piece of their planned deployments. With 80%-85% of retail data never leaving the store, these applications, and many others, demand that AI processing move closer to where this sensitive data is generated. Moving data to the cloud for processing is not always a viable option for speed, cost, or privacy reasons. As such, AI in retail will become increasingly "hybrid," combining inference and eventually training at the edge with that from the cloud. This allows constant interaction across various model types, latency needs, and regulatory restrictions, and provides a rich combination of deep insights for retailers.
There are several exciting use cases emerging. These include enabling customers to skip checkout and simply leave the store, robots that pick orders at the warehouse, smart shelves that measure inventory in real time, computer vision to understand store traffic patterns and smart kiosks to provide self-service options for customers. By automating most of the transactional interactions, employees can focus on helping customers and other high-value tasks. Common use cases of AI at the Edge in retail will include:
Frictionless shopping and checkout
Perhaps the primary use case is utilizing AI to streamline shopping and checkout. Many shoppers prefer the convenience and speed of doing things for themselves by using their phones or a technology service. Deep learning, computer vision and interface technologies that are highly compelling and engaging are needed to deliver offerings like "Smart Shelves," contextual promotions and even eliminating checkout. Freeing up employees from mundane tasks such as manual checkout enables them to provide more value to shoppers, who of course benefit with faster checkout. Innovative retailers will also use these systems to personalize recommendations, promotions, and customer engagement.
Loss prevention
Store theft remains a big problem for retailers, many of which have found that loss prevention can cost more than the lost goods are worth. The ability to reduce losses without an increase in costs is extremely attractive. That is what happens when computer vision is combined with AI. For example, object detection at self-checkout stations and motion analytics performed at the edge can identify potential loss and enable staff to respond in real time. The computer vision-AI combination can also anonymously surveil store traffic and identify potential loss scenarios as they occur. Some retailers have already deployed these systems and are using edge infrastructure to reduce losses before they happen.
Personalized customer experiences
Personalization has been a retailing priority for some time, but retailers today are looking for ways to tune personalization to a specific visit or “in-the-moment” needs of customers. That takes a lot of data, making AI at the edge essential to ensure that customer experiences occur at human speed. Retailers are also experimenting with digital signage and other devices to augment the personalized experience or recommendations, using local AI infrastructure to analyze past purchases, their current path in the store, and other inputs in real time. Another possibility is to use AI to analyze the path that a customer takes in a store and then provide location based contextual recommendations or offers depending on what the customer looks at.
Smart Retail Solutions Deliver Seamless Checkout
Shoppers are now accustomed to fast, hassle-free ecommerce checkouts and now expect a similar experience in physical stores. Learn how new self-service kiosks have improved immensely with AI software that can recognize hard-to-scan items and accepts a variety of payments.
Download NowInformational interactions with conversational and generative AI
Enabling customers to have an interactive conversation with a truly intelligent artificial agent provides a number of benefits. The fast-food industry will use this technology to speed up and improve order accuracy at drive-throughs while enabling upsells. In stores, many simple queries will be resolved quickly and accurately without having to wait for store staff. Conversational AI will also help retailers better determine what store experience fits best and the items of interest, as well as to provide real-time inducements. This technology can also provide immediate assistance for locating items in larger stores.
Intel and its partners support retailer deployments of AI in the store
Retailers that can best leverage AI and the edge will be winners in the market. Leveraging a consistent, proven, and cost-effective infrastructure for AI at the edge will be essential to these and many other use cases. Many retailers believe that new infrastructure is needed for AI. However, with thorough evaluation, they’re discovering starting from scratch and trying to scale that from proof-of-concept to full deployment is prohibitively expensive or complex. Retailers that have more mature AI deployments are finding they can use their largely PC/Intel-based infrastructure to run many AI solutions. Whether it’s a small boutique or a multinational superstore, the Intel portfolio provides a strong foundation and lowers the barriers to entry for retail AI. Intel has supported tens of thousands of edge AI deployments and has deep expertise in this space. Unique software like OpenVINO unifies edge-to-cloud inference and streamlines the development of real-world applications. Hundreds of market-ready solutions are available from their world-class ecosystem. For more information, please click here.
Notices & Disclaimers
Intel technologies may require enabled hardware, software or service activation.
No product or component can be absolutely secure.
Your costs and results may vary.
Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.
