Don’t React Hastily to GenAI: Considering the Customer and Employee Experience

Big Hype or Big Opportunity?

The hype around generative AI has snowballed. Gartner has artificial intelligence pegged at the peak of hype in its latest Artificial Intelligence Hype Cycle. And the writing’s on the wall: they also predict that in two short years, 30% of outbound marketing messages from large organizations will be synthetically generated, up from less than 2% in 2022. 

GenAI has captured everyone’s attention—every CX leader we work with across industries: financial services, retail, communications, logistics, healthcare—has GenAI at the top of their agenda. While it may feel like the hype surrounding the metaverse last year, GenAI is different, a lot more substantial, and on the path to becoming one of the most disruptive technologies of our time.

Our take: The benefits are real. Generative AI is rewriting the rules of customer and employee engagement. But so are the risks. GenAI is a powerful tool for creating owned brand experiences at the intersection of data and technology that work for customers, employees, and businesses. So, what is the right way to explore GenAI’s power and potential to create meaningful outcomes?

Early Successes and GenAI-Powered CX Applications

Recently, Adobe surveyed 4,250 customer experience and marketing professionals across 14 countries and discovered that marketing and customer experience leaders believe generative AI will help in various ways.

Some big companies have already implemented GenAI solutions to enhance user engagement and customer experiences, such as Spotify’s AI-driven DJ and Coca-Cola’s Create Real Magic platform. Design tools have also made generative AI part of their process. For instance, Adobe Sensei generates design variations through user input, including customer preferences, to create images more likely to lead to a sale. Wireframe Designer, a Figma plugin, uses ChatGPT3.5 to automate mobile wireframe designs, enabling designers to save time while developing more iterations that lead to better design. 

Yes, you probably see a pattern—these companies are the global leaders in their space, with deep pockets and reach. With their big budgets and international muscle, global leaders like Microsoft and Amazon are the ones integrating GenAI into many of their tools and services—because they have been able to invest in the space for years, and they still have not fully worked out the kinks before unveiling them publicly.

GenAI: Caution Ahead

However, as a CX leader, you know better than to jump into the GenAI frenzy without a plan. So, before you sprinkle your CX with GenAI pixie dust, you should first understand where it can have the most impact. You also sense the risks of not doing it right may outweigh the initial enthusiasm. Indeed, you are on the right path. 

The most common mistake with GenAI is to focus it on automation rather than augmenting human decision-making and improving customer interactions. So, you should look for critical business points where human interaction or expertise adds value. The truth? While generative AI has made big promises, it's not yet mature enough for many organizations to fully deliver on them.

According to the recent CX20 Global Study from Stellar Elements, CIOs, CTOs, and other IT decision makers ranked the AI Readiness Gap as their top gap to address.

Moreover, there are many risks and issues when using the technology. Recently, The Harvard Business Review demonstrated that generative AI-based tools can suggest incorrect recommendations or introduce code errors. Generative AI can “hallucinate” when it uses training data to construct entirely new data based on the patterns it learned. This false yet convincing data can mislead people into believing in false facts.

You can’t trust GenAI completely. It’s not a magic box. It’s pushed beyond autocomplete to a new threshold into something else. Your keyboard’s autocomplete on steroids.

Training foundational models is costly, but tweaking and experimenting with foundational models is not. Lower cost options are to use what’s already out there—but companies have privacy concerns about their data. 

Reid Blackman, the author of “Ethical Machines,” rightfully pointed out, “there is also well-founded anxiety.” Among others, Samsung banned using ChatGPT after employees loaded sensitive company data onto the leaked platform. Additionally, Blackman shows the well-documented tendency of AI to generate discriminatory outputs applies to generative AI, including a slew of generative AI companies that are facing lawsuits. The AI image generator, StableDiffusion, faces a lawsuit from Getty Images, while Microsoft, GitHub, and OpenAI also confront a class-action case. 

Despite these issues, generative AI will mature quickly to widespread adoption across the customer and employee experience. Let's look at where it is already making some substantial CX headway.

Here’s how to get there.

Build an honest self-assessment of where you are with your customers and how they’ll interact with you in the future.

To deliver outcome-driven innovation that can transform your EX or CX with immediate and lasting impact, you and your team must connect the dots between your customers’ experiences to fully understand the holistic one. These opportunities can be varied. GenAI can improve your business by streamlining processes, automating support, enhancing insights, or personalizing interactions.

For example:

1. Conversational Implementations—The inherent ability of these large language models (LLMs) to interpret natural language inputs with a high degree of accuracy and flexibility enables solutions that humans can easily interact with. These solutions often outperform traditional intent or algorithmic input methods with significantly fewer data requirements.
   

2. Generative Focus—The massive amount of data these models are trained on and retain embedded in their weights amplifies LLMs beyond just "next best word" predictors. They can expand, compress, transform, and extrapolate complex natural language.
   

3. Synthetic Databases—With careful prompting, you can manipulate these models into generating unique and relevant datasets. These can help in training simulations for staff, autonomously interact to identify areas of opportunity within the business or enable autonomous and informative reporting and resolution.

Resist the urge to jump headfirst into GenAI. First, sharpen your CX pencil and unearth the CX or EX opportunities with the most outcome promise or innovation impact. Then, you can safely point your GenAI guns at addressing the specific outcomes and results your employees and customers want to achieve with powerful precision.

More About Stellar Elements

A global design and development firm with a 20+ year track record of delivering and scaling connected customer, employee, and partner experience solutions for organizations in the telecom, financial services, retail, technology, quick-service restaurant (QSR), and healthcare industries, we answer the most critical and essential questions for our clients in every corner of the world. Our customers include EE-BT, Globe Telecom, Capital One, Mercedes-Benz, Dell, and 50+ Fortune 500 companies.