A survey published today finds that while there is a significant amount of dissatisfaction with chatbots, interactive voice response (IVR) services and conversational assistants, enthusiasm for the ChatGPT platform among digital quality testing professionals is a lot higher.
Conducted by Applause, a provider of a platform for accessing a networking of independent testers, the survey of 5,200 digital quality testing professionals finds nearly one-third (30%) of respondents are dissatisfied with their current chatbot, IVR and conversational assistant experiences, with wrong answers (29%) and a lack of understanding (24%) as the top two reasons cited.
Nearly three quarters (74%) said they had a positive experience with the ChatGPT generative artificial intelligence (AI) platform, with more than two-thirds (67%) reporting the platform provided helpful answers to questions. In terms of work-related tasks, most said ChatGPT would either be “very” (43%) or “somewhat” (37%) helpful.
The survey results suggest that enthusiasm for ChatGPT as an alternative approach for automating customer service and support is surprisingly high, says Luke Damian, chief growth officer for Applause. “Most digital quality testing professionals tend to be dubious of cool prototypes,” he added.
Despite current levels of enthusiasm, however, the survey also finds 86% of survey respondents are concerned about biases that might be baked into the ChatGPT platform. As such, many expect to continue to invest in other AI technologies, with 92% of respondents expecting companies to have chatbots in their applications or websites, with 44% expecting mobile applications to have voice assistants. A full 86% said they also expect to be greeted by an IVR system when they call a business.
It’s not clear how customer service and support systems might invoke ChatGPT but there is an application programming interface (API) that can be used to launch ChatGPT prompts from within an application. Some organizations may conversely opt to employ a generative AI platform that has been specifically trained on a corpus of data that is applicable to their vertical industry. Regardless of approach, it’s clear that once trust is established next-generation AI platforms are going to have a profound impact on how customer service and support is delivered. “It will be pervasive,” says Damian.
Trust, of course, is easily won and lost. It only takes a few wrong answers for a user of a platform to become wary of everything else an AI platform might present. Organizations that employ generative AI without first validating the results are likely to be subject to social media ridicule, so caution is advisable.
Nevertheless, no organization wants to be left behind once a specific technology becomes a capability that is routinely expected. The issue now becomes what role existing customer service and support professionals play as more issues are handled by generative AI platforms. There will always be complex problems that require human intervention, but most routine issues might be easily resolved via a ChatGPT prompt. The challenge, of course, is making it simple as possible to find the right answer without requiring an end user to have advanced prompt engineering skills to interrogate a generative AI platform.