AI SERVICE
CONTINUED
are language rules pertaining to verbiage, emotional tone, perspective, adjective use and clause length during the construction of every filter. It takes a writer to explain Latinate vs Anglo-Saxon vocabulary, or the intricacies of the active voice, or the difference between abstract and concrete nouns. “No one wants to sound the same and given the chatbot market is predicted to enjoy a CAGR of over 19% in the next decade, this problem is only going to be exacerbated. In January the former director of AI at Tesla, Andrej Karpathy, tweeted “The hottest new programming language is English“, to be honest, I thought, ‘Yeah, good one’ at the time, given naysayers predicting the end of all human language output in every form because of GenAI, but ironically, given the demand we’re seeing for our language team, it would appear Karpathy was on to something.” Coming to Customer Although the idea of an artificial human voice may be disconcerting to some, it may have the opposite effect on the vast majority of customers, as brands become more relatable. According to Rafferty, this may not necessarily be in the form of imitating real people, instead, guide agents to the best solution. “The huge rise of AI continues to shape the way businesses deliver customer service. AI now stands at the forefront of human- machine interactions, accessible to both the broader workforce and consumers alike. Not only can AI-powered solutions automate time-consuming tasks, such as updating CRM logs and sending follow-up emails to free up agents, but AI can also assist with direct customer interactions, guiding agents on what to say to improve and personalise customer support experiences.” “Businesses using AI should enhance self- service tools to help speed up the customer service process. Usually, customers just want to get help quickly without waiting on hold for a live agent or having to navigate complex menu options. With AI, chatbots can better provide intelligent instant responses to simple customer queries or automatically gather information upfront before transferring to a live agent. Using AI to prioritise customer support tickets is a great way to ensure customers get the service they need when they need it.”
Great customer
service doesn’t change just because we’ve found a new tool to add to our already extensive kitbag.
Installing into Business Whilst the capabilities are intriguing, and potentially timesaving, installation is one of those hurdles we all have to come up against. We all know the statistics about ChatGPT adoption, and saw the embarrassing headlines that followed. As a result, Morton says that businesses should have a particular problem that needs solving in mind before looking to add AI to the CX stack. “Let’s not forget how AI actually works, as well as the importance of being cognisant of public and private data sources, both of which are vitally important aspects of how we can deploy AI. It is my prediction that clients will look to ‘private AI’ solutions and, as such, ‘AI in a box’ will become a service offering in the market. “What AI is great at is intuitively sourcing data. But what it’s not so good at is nuance, tone, sentiment, empathy or brand value. Great customer service doesn’t change just because we’ve found a new tool to add to our already extensive kitbag. Customer journey mapping will still be as relevant tomorrow as it was yesterday. “As such when deploying any new tools, including those which are eminently intuitive and compelling, remembering what problem we are trying to solve, how the solution will address the customers enquiry, and how do we want our brand to be portrayed, must remain front of mind and not become secondary to the latest technology.”
Steve Rafferty VP International
ringcentral.com
20
Powered by FlippingBook