ChatGPT and Customer Service: A Breakdown of New Tech's Impact on CX

ChatGPT is having a moment. After its release this past December, users have sung its praises, highlighting its unique functions, creative responses and human-like voice. With examples flooding the internet, ChatGPT conversations have attracted a viral appeal — any new and clever response is met with an abundance of acclaim. The technology has also simultaneously prompted a discussion of the ethics of relying on intelligent computer generated writing.

But, before diving into the backlash, let’s break down the technology itself. 

What is ChatGPT?

ChatGPT is an AI-powered chatbot that uses natural language processing to generate conversations based on user questions and prompts. The technology allows users to provide specific prompts and answers with in-depth insight on almost any topic. 

This is not your typical one-track chatbot — ChatGPT can respond with answers to legal questions, generate and explain code and even pass a final exam at the Wharton School. The technology has high caliber capabilities and its being considered an influential new tool akin to the iPhone and Google search. 

To understand its current reach, Ryan Reynolds, actor and owner of Mint Mobile, used the tool in a recent commercial for the brand. In the clip, Reynolds asked ChatGPT to write a commercial in the voice of Ryan Reynolds. He told it to use a joke, include a curse word, and let people know that Mint’s holiday promo is still going even after the big wireless companies have ended theirs.

Here’s how it responded: “Hey, it’s Ryan Reynolds here. First of all, let me just say Mint Mobile is the shit. But here’s the thing, all the big wireless companies out there are ending their holiday promos, but not Mint Mobile. We’re keeping the party going cause we’re just that damn good. Give Mint Mobile a try. And hey, as an added bonus, if you sign up now, you’ll get to hear my voice whenever you call customer service. 

Just kidding. That’s not really a thing. And stay classy, everyone.”

Reynolds noted that this accurately crafted paragraph is “mildly terrifying, but compelling.” And, it seems this sentiment is largely echoed across industries.

ChatGPT is Not all Fun and Games

While many have certainly had their fun with the tool, others are markedly concerned with its overarching abilities. Leveraging its eloquent responses as a means of bypassing genuine work and creativity, the technology has already opened up opportunities for misuse and deception. 

The New York Times already reported on a case of plagiarism at Northern Michigan University. Antony Aumann, a philosophy professor at the University, flagged “the best paper in the class,” on the morality of burqa bans. The essay apparently boasted clean paragraphs, comprehensive examples and rigorous arguments. 

The professor immediately recognized a disconnect between the student’s previous work and confronted them. The student later confessed to using the ChatGPT chatbot, beginning an alarming trend of plagiarism and exploitation of the intelligent technology.

Now, educators are taking steps to discourage students from engaging with the technology, making changes to include more oral exams and handwritten assignments. By avoiding entirely digital assignments, or including more transparency in the writing process, educators are hoping to safeguard creativity, personality and critical thinking skills in the classroom.

Edward Tian, Founder of GPTZero, a tool designed to identify the use of ChatGPT in writing, shared his larger concerns over the loss of personal voice and creativity ChatGPT could potentially cause. He states, “Now humanity faces the prospect of an even greater dependence on machines. It's possible we're heading towards a world where an even larger swath of the populace loses their ability to write well. It's a world in which all of our written communication might become like a Hallmark card, written without our own creativity, personality, ideas, emotions, or idiosyncrasies. Call it the Hallmarkization of everything.”

However, this goes beyond educational settings. Concerns over ChatGPT in the workplace have simultaneously come to light. 

In a recent experiment, one business owner decided to test whether ChatGPT generated writing could impress his own recruitment team. To get a better idea of how worried he should be about the technology, Neil Taylor entered ChatGPT’s writing in an application for an open role. With the writing brief, "In 300 words, tell us the secret of good writing," ChatGPT was able to secure a first round interview with the consultancy firm. With less than 20% of applicants shortlisted,Taylor confirmed that it outperformed many of the other ‘bad’ applicants. However, it did take multiple rounds of tweaking with the AI chatbot to achieve the end result.

ChatGPT and CX

Anyone who has engaged with an ineffective chatbot will instantly recognize the benefits of a more conversational and creative chat experience. Users who have been stuck in an endless loop with a chatbot, never reaching a resolution, or a human, will likely look forward to the day that brands could leverage this technology. But it seems it’s not quite there yet.

According to ChatGPT CEO Sam Altman, the technology “is incredibly limited, but good enough at some things to create a misleading impression of greatness.” He furthers in a Twitter statement, “It's a mistake to be relying on it for anything important right now. it’s a preview of progress; we have lots of work to do on robustness and truthfulness.”

One of its best features is also its current downfall — the writing appears highly proficient and the voice is incredibly confident, but when the content behind ChatGPT’s robust responses is not truthful, it can be difficult to pinpoint misinformation. Therefore, in its current state, ChatGPT acts as merely a resource. Users must be knowledgeable enough about the given topic to recognize factual inaccuracies.

Should, or could, it be used in the customer service space, then? In terms of delivering better digital experiences, it is promising. This technology could seemingly be a stepping stone toward more conversational, intuitive and interesting chat experiences. It can also be a ‘behind the scenes’ tool for agents looking for more guidance on delivering comprehensive, compelling responses. 

However, currently, with factual inaccuracies it could not be used effectively as either. As their CEO states, the technology can confidently mislead users. Part of its appeal is its unabashed confidence and ability to create largely human and highly intelligent responses. If the user is unable to decipher the truth, they could be spreading false information and ultimately disappointing or misleading customers.

At the end of the day, ChatGPT will continue to grow, leaving room for continuous improvement. Whether or not it will change the way we write, work or respond to customers, it will certainly impact the future of chat experiences. As we look to implement change in the contact center, we will undoubtedly look to tools like these to continue engaging and exciting customers.

For more discussion on ChatGPT and the future of experiences, join Christine Ducey, Content Analyst and I as we break down the top technology do’s and don’ts during the CCW San Antonio Presents: State of Contact Center Technology Event.

Check out our latest agenda and register here.

Front image by Ketut Subiyanto

Second image by Jep Gambardella