Generative artificial intelligence is making waves across the retail industry, from consumers using the technology to inspire purchasing decisions to retail brands employing chatbots to improve customer service and experience.
While chatbots have been around for quite some time, with the advent of generative AI, there’s been a huge revolution over the last 18 months to 24 months, particularly with the launch of ChatGPT.
Why? The outcome from generative AI chatbots can be even more impactful than those composed from classic AI. So how can retailers integrate generative AI to positively affect their business this holiday season and beyond?
To understand how to safely and effectively employ generative AI chatbot elves, Santa created a chatbot checklist for retailers to land a spot on his nice list for the coming holidays:
- Build chatbots with prompt engineering that include specific safeguards. Just as people can get distracted and go off topic from the task at hand, chatbots can do the same. Fortunately, there’s a solution to this through prompt engineering. This practice allows humans to train and implement guardrails that keep chatbots on the right track and focused on the customer's needs.
- Use classical AI to fact-check and correct generative AI chatbots. Despite taking all necessary precautions, mistakes can still happen. Generative AI chatbots can be kept in check by other AI systems that double-check their answers and flag questionable responses that may contain sensitive information or inappropriate content before they're sent out.
- Continuously optimize your system. The chatbot journey isn’t complete once the solution has been fully integrated into your company. It’s crucial to continuously monitor chatbots for any major issues or complaints from customers, ensure accurate information is being released and that it reflects the corporate brand, and, most importantly, keep learning about them. Following these steps can help assure consumers will get accurate and efficient help through the use of chatbots.
And what not to do, unless you want to be on the naughty list:
- Don’t connect chatbots directly to backend systems. Generative AI has the unique ability to craft its own responses, without a script, based on the information it’s trained on. Therefore, be careful what information you give it. If a chatbot goes directly to the backend system, it risks pulling confidential or sensitive information that shouldn't be released to users. Instead, companies should have chatbots take in customer requests, provide the input to normal middleware, and then have this middleware make the request to the backend system, returning only information that you would be happy to share with your customers.
- Don’t overlook the value of user feedback. The top priority of customer service is ensuring customers are satisfied with their experiences. This holds true across all facets of customer experience, from live agents all the way to chatbots, making consumer feedback the key to a successful interaction. By using feedback, companies can better understand what makes their customers happy and what’s needed to improve satisfaction.
- Don’t over-rely on technology; incorporate human oversight. Technology, much like humans, isn’t perfect and can make mistakes. It’s necessary to have another set of eyes monitoring work to ensure that everything is running smoothly and that it’s reducing the risk of security breaches or incorrect responses. Despite advancements in technology, humans are still the best answer in this case.
To make the most of your own chatbot elves this holiday season, take your customers’ best interests into consideration and aim to provide a safe and secure experience. It’s the most wonderful time of the year, after all.
Kerry Robinson is vice president, conversational AI at Waterfield Tech, a contact center solutions provider offering a full suite of services and platforms.
Related story: How Retailers Can Use Conversational AI to Drive Positive CX This Holiday Season and Beyond
An Oxford physicist with a Master’s in Artificial Intelligence, Kerry is a technologist, scientist, and lover of data with over 20 years of experience in conversational AI. He combines business, customer experience, and technical expertise to deliver IVR, voice, and chatbot strategy, and keep Waterfield Technologies buzzing.