What are the short-term implications of ChatGPT for CX?
by Ray Poynter
Before we delve into ChatGPT and its implications for CX, here is a quick intro to what it is….
ChatGPT is a conversational AI model developed by OpenAI. It is specifically trained to generate human-like responses in a conversational context and has been fine-tuned on a large mass of conversational data. It is designed to generate human-like text based on the input prompt given to it, allowing it to engage in a dialogue with users and answer their questions in a natural and informative way.
In the medium to longer term, the implications of Generative AI for CX are massive and transformational. But, what about the short-term impacts of ChatGPT, the first Generative AI system to break through into generalized, public awareness?
The Blurring of DIY and Service.
In my recent 2023 CX Trend Report, I talked about the blurring of DIY and service. ChatGPT is already playing a role in this. Service teams are using ChatGPT to review emails and to generate responses, or to suggest edits to responses they themselves have typed. The next step will be to utilize the apps associated with ChatGPT to get the AI to respond directly (and almost immediately) to enquiries from customers.
ChatGPT can write code, and this may help teams creating systems. However, ChatGPT can also create realistic consumer questions and dialogues, allowing developers to use ChatGPT as a testbed for new ideas and systems.
ChatGPT can be asked to help create scenarios for staff to role-play, to improve their training. Sales staff can role-play with different types of customers. Service teams can explore different ways of dealing with a series of problems when presented by different types of customers.
The first problem is that ChatGPT makes mistakes, as it freely admits. This might mean sending the wrong message to a customer (although humans already do that too) or that the code it writes does not work properly (and humans already do that too).
ChatGPT could be used to try to game CX systems, to jump queues, to get compensation, or even to harass people. Customers may struggle to know if they are talking to a person or a generative AI bot. Similarly, brands may also struggle to know if they are talking to a customer or a generative AI interface.
One specific form of cheating we might see is the use of ChatGPT to create fake reviews. These fake reviews could be designed by brands to promote their products and services, or to attack competitors, or by third parties simply to create mischief.
ChatGPT and Employment
If we are talking about 2023, then ChatGPTis unlikely to replace people, it is too new, too risky, and it will take additional resources to set it up. It is more likely that ChatGPT will increase productivity and reduce response times. However, in the longer term, it seems likely that generative AI will increase the level of service and at the same time reduce the number of people required to deliver that service – just as ATMs improved our ability to get cash AND reduced the number of bank tellers.
What does ChatGPT think?
I asked ChatGPT “What are the best ways to use ChatGPT in the area of customer experience and CX programs?”. It said:
“There are several ways ChatGPT can be used in the area of customer experience (CX) and CX programs:
1) Customer Service:ChatGPT can be used as a virtual assistant to answer customer queries and provide support in real-time, improving the overall customer experience.
2) Feedback Collection: ChatGPT can be integrated into customer feedback systems to gather insights and opinions from customers, which can be used to improve CX programs.
3) Chatbots: ChatGPT can be used to build conversational chatbots that can handle a wide range of customer interactions, such as booking appointments, providing product information, and answering frequently asked questions.
4) Personalization: ChatGPT can be used to personalize customer interactions by using information such as customer history, preferences, and context to provide relevant and tailored experiences.
5) Data Analysis: ChatGPT can be used to analyze customer data and provide insights into customer behavior and preferences, which can be used to improve CX programs.
Overall, ChatGPT can be a valuable tool in improving the customer experience by providing quick and effective support, collecting valuable feedback, and offering personalized interactions.”
How Short is the Short-term?
I suspect the short-term will be very short, with 2023 seeing a raft of new changes. For example, Microsoft has announced that they are linking ChatGPT with their search engine Bing. Google has announced the release of their Generative AI tool, called Bard. Amazon is working with Stability AI to produce options in this space too. CX will need to move fast to keep up to date with these changes.
The Big Picture?
Generative AI is going to create significant changes for how CX is delivered and measured, and the rate of change is going to be fast. I recommend that organizations start conducting small-scale experiments immediately, to discover what is possible, and to identify potential opportunities and threats.