Generative artificial intelligence (AI) platforms like ChatGPT are the latest sought-after technology among retailers and the supply chain.
While conversation technology has been around for several years, it has recently become a more significant investment for businesses like Walmart and Amazon, which have already started applying chatbots to their business relationships.
A December report from Harvard Business Revenue indicates Walmart seriously began to deploy a generative AI chatbot product from Pactum AI in 2021 after a 2019 test in the Canada business showed promise.
Walmart International experimented with the chatbot to see if it could close deals with 20% of its suppliers in the 2021 pilot. The task called for the chatbot to negotiate with suppliers for items used by Walmart, like shopping carts, fleet services and other store equipment.
Also, as part of the trial, Walmart targeted payment schedules hoping to negotiate early payment discounts or extended payment terms without discounts. As a trade-off, Walmart offered suppliers the option to change the retailer’s right to terminate the contract immediately to an option that gave 30, 60 or 90 days written notice.
Walmart also offered suppliers opportunities for growth in assortment and sales volume in exchange for price discounts. The chatbot did all the negotiating after being trained by Walmart merchant teams.
Walmart said 89 of the 100 suppliers invited took part in the pilot. The chatbot reached an agreement with 64% of them, three times the target of 20%. Walmart said it gained an average of 1.5% in cost savings and an extension of payment terms to 35 days.
The report said 83% of the suppliers liked the chatbot negotiation. However, several also said they would have wanted to negotiate face-to-face. Walmart considered the pilot a success and has since expanded to its not-for-resale suppliers in the U.S., Chile and South Africa. Walmart said the chatbot had closed deals with 68% of the suppliers approached and generated an average savings of 3%.
Walmart has continued to use chatbot technology in other areas of its business. The retail giant said its text-to-shop service rolled out in December has already resonated with 50 million customers. Text-to-shop uses chatbot technology that allows customers to text the items they want into the Walmart app and have them automatically added to their online cart.
Internally, Walmart uses a chatbot to help employees get their desired answers. The “Ask Sam” chat function answered 1.3 billion questions last year by Walmart employees, Walmart CEO Doug McMillon noted in the company’s annual report.
“More than ever, we’re being powered by technology. Our customers and members engage through our apps and sites, and our associates have information and capabilities at their fingertips. We’re learning to put our data to work,” McMillon said.
ChatGPT and other intuitive chatbot software can be adapted to the supply chain in at least five applications, according to a report from the trade magazine “All Things Supply Chain.”
Predictive analytics software can gather and study data from different sources, such as customer orders, production schedules and inventory levels to better predict future demand and supply. Many companies already use predictive analysis up and down the retail supply chain because it helps increase efficiency and reduce waste.
Customer service is another way retailers have begun to use chatbots. Justin Richie, vice president of data at Bentonville-based RevUnit, said many consumers had had chatbot conversations while shopping online. He said this level of rules-based machine learning had been the entry point for many companies because it helps them accomplish tasks like answering routine questions. Richie said the experience is only sometimes the best for the consumer.
“With generative AI and the advancements in large language chatbots like ChatGPT, this experience will improve over time. It will seem more like you are talking with a person,” he added.
Inventory management is another way retailers and suppliers use ChatGPT to help them monitor inventory levels in real-time and better predict when specific products will run out of stock, helping to avoid costly out of stocks, according to “All Things Supply Chain.”
Content generation is also popular for retailers to use chatbots to generate blog posts based on their data or a more expansive universe of third-party data to create content used in sales and marketing. Consumer products giant Unilever said it uses generative AI across its diverse businesses. One application is a system Unilever calls “Homer” that leverages ChatGPT for content generation. It’s a neural network that takes a few details about a product and generates an Amazon product listing with short and long descriptions that match the brand tone.
“We want to ensure we capture the voice of the brand so, for example, that we differentiate between a TRESemmé and a Dove shampoo, and the system got it absolutely nailed,” said Alessandro Ventura, chief information officer at Unilever, North America.
Unilever also said it’s using another AI-based tool to support the Hellmann’s mayonnaise brand in its goal to reduce food waste. That initiative launched in November.
“It links up with the recipe management system that we have at Hellmann’s, so somebody can go in and select two or three ingredients that they have in the fridge and get in exchange recipes for what they can do with those ingredients,” Ventura said in the release.
Richie said there are many applications for retail and the supply chain. He said chatbots could provide companies with more time as they are best at handling redundant tasks. He can foresee a time when chatbots become more of a change manager, allowing businesses to get the best use for their AI investment. He does not see chatbots replacing humans but helping humans increase productivity.
As technology usage rises, so will the threats of hackers and bad actors. Richie said companies could protect themselves by siloing the data search area only to include their wholly owned data and content.
He said that as more technology firms get into this highly competitive area, retailers should seek solutions that best align with the task and objectives they are trying to accomplish. While ChatGPT is not free, other open-source avenues will likely accelerate the adoption curve and the risks for hackers.
“Companies that use ChatGPT and other chatbot systems must also set firm guidelines with their staffs to ensure private or proprietary information is never shared with the bots,” Richie said.
Walmart issued its guidelines in February as global tech employees were told the retailer had noticed ChatGPT activity that presented a risk. Walmart said the new guidelines do not allow employees to input sensitive, confidential or proprietary information such as financial, strategy or personal shopper information into ChatGPT. Employees were also told not to cut and paste existing code into the tools to create new code as it would raise the risk of breach.
Amazon and Microsoft have similar employee guidelines in place for ChatGPT and generative AI chatbot technology.
Editor’s note: The Supply Side section of Talk Business & Politics focuses on the companies, organizations, issues and individuals engaged in providing products and services to retailers. The Supply Side is managed by Talk Business & Politics and sponsored by Propak Logistics.