Enterprise technology is evolving at an unprecedented speed, and AI is the rocket fuel.
Yet, customer support teams have little experience with AI beyond previous-generation chatbots, ChatGPT experiments, and what's embedded natively into their everyday systems.
As such, keeping up with the innovation and learning curves may seem too tricky.
Thankfully, CX Today’s All-Stars have kept their fingers on the pulse, sharing top takes on the future of AI in customer support to help bring their fellow practitioners up to speed.
Here are seven of the most striking predictions.
-
CX Teams Will Start to Leverage Large Behavior Models (LBMs)
Contact center providers have developed natural language processing (NLP) models to detect customer sentiment for years.
Now, LLMs are helping this capability progress, enabling live and virtual agents to detect specific moods, adjust their tone, and deliver experiences that feel human-centered and empathetic.
Large behavior models (LBMs) are the next frontier for these use cases, and – according to Michael Fauscette, Founder, CEO & Chief Analyst at Arion Research – they’ll begin to underpin emotional CX.
"LBMs, unlike the more widely used LLMs (and variants like small and medium models and industry-specific models), are trained on human behaviors instead of text," he noted.
"This creates the ability for the model to recognize human behavioral cues and accurately predict the best actions to take based on those cues and predicted behaviors."
As such, conversational AI vendors may leverage this technology to power smarter, multimodal virtual agents – or perhaps even avatars.
-
Predictive Metrics Come to the Fore
The death of NPS surveys is a perpetual prediction that comes back every year.
There are plenty of reasons why they should bite the dust. Low response rates, customer fatigue, and twisted calculations… take your pick.
Yet, NPS surveys are still here. There are two key reasons for this:
- Some business leaders have become NPS obsessed, chasing the score.
- A more compelling alternative hasn’t come along.
Thankfully, predictive metrics are bubbling to the surface to overcome the issues above.
As Luke Jamieson, a CX Evangelist, observed: “Surveys will become predictive, based on language, subject, tone, and effectiveness of interactions.”
Indeed, solutions that predict NPS scores – powered by generative AI – are already on the CX market. They evaluate each customer support interaction to decipher a customer's score, should they have completed a survey.
That brings several possible benefits, as outlined in the article: Predictive Metrics Will Help Shape the Future Contact Center. Here’s How.
-
Transactional Interactions Become Part of Something Much Bigger
The biggest shift in the future of CX is the transition from transactional experiences to relational ecosystems. That’s according to Justin Robbins, Founder & Principal Analyst at Metric Sherpa.
“Customer interactions won’t be standalone events; they’ll be part of a continuous, interconnected journey powered by AI, data, and human ingenuity,” he added.
To start that journey, many businesses pool their customer support, sales, and marketing data into a customer data platform, like Salesforce Data Cloud or Twilio Segment.
From there, brands create a more rounded view of the customer, funnel new data into AI models, and give customer-facing employees more of that all-important context.