The AI revolution has ushered in sweeping changes for retailing, transforming national chains, local franchises, and independent operations alike. Especially in the supply chain, the underlying technology has helped streamline supply-side operations, reduce overhead costs, and achieve more precise inventory management. But there’s no reason to expect that the changes will stop there, as a recent innovation by Scotmid, Scotland’s largest independent cooperative, indicates.

In a partnership with Retail Insight, Scotmid has developed and implemented technology designed to make a huge difference in the critical, global problem of food waste. Specifically, Retail Insight provided various Scotmid locations with a pricing tool that keeps close track of inventory, then suggests price reductions as goods get closer to their expiration dates. This data-driven model also monitors consumer buying patterns, to prevent overstocking or understocking situations. Thus far, its implementation already has resulted in a 1.83 percent increase in products sold, as well as a reduction of approximately 42.7 tons of food waste.

Such promising outcomes seemingly suggest that similar tools should be added to every food retailer, everywhere in the world. But challenges to implementing AI inventory management systems remain. In particular, new information systems can be unwieldy and difficult to integrate with existing systems. Store associates also tend to express resistance to learning about new technology. Data inaccuracies have been reported with some programs, including faulty buying predictions that lead to inventory mismatches and consumer complaints. Consumers also express concerns about the use of AI technology in stores, citing their desire for privacy, which thus far is not protected by any clear regulations.

In response to such concerns, some retailers have purposefully tried to reduce the scale of their data management systems, focusing on finding solutions to specific challenges, including food waste. But the challenges also indicate the likely need for continued human oversight. Even if AI systems can be used to enhance inventory management and manage overhead cost, they remain unable to replace certain human tasks. Ideally, retailers might adopt a partnership model, adding new systems with the assistance and collaboration of existing staff, to balance human and machine fallibilities.

Finally, large retailers might be adopting AI tools to reduce their contributions to food waste, but significant barriers remain for smaller businesses, even if they want to add the expensive and relatively complex technologies to their operations. Considering the global relevance of food waste and the pressing need to mitigate its detrimental effects, there may be an argument for making AI dedicated to this particular outcome more freely and readily available, throughout retail food markets.

Discussion Questions

1. Do you think about food waste when purchasing groceries?
2. Can AI and related technologies be classified, into those that provide social benefits (and thus should be widely available) versus those that only benefit the retailer (and thus should be subject to conventional purchasing processes)? Would doing so be appropriate?

Sources: Andrew DeLanzo, “The AI Revolution in Retail: Transforming Marketing, Inventory, and Customer Experience,” AI Time Journal, October 14, 2023; Dan Macnamara, “Scotmid Co-op’s AI-Driven Strategy to Minimize Food Waste,” Retail Tech Innovation Hub, March 11, 2024; OpenAI ChatGPT, “Assistance with Research on AI-Driven Inventory Management in Retail,” April 22, 2024; Rudrendu Kumar Paul and Apratim Mukherjee, “Harnessing AI for Retail Supply Chain Optimization,” Retail Technology Review, August 4, 2023