By Jordi Torras, Chief Innovation Officer at Inbenta
In just a few months, ChatGPT has captured the imagination of hundreds of millions of users and businesses around the world. The meteoric rise of the Generative AI platform is understandable. a new and transformative AI tool capable of debugging code, answering questions, writing emails letters and articles, and everything in between, are now widely available to the public.
The data used to train ChatGPT is also impressive: 45 terabytes of text data. ChatGPT estimates that’s the equivalent of about 30 billion e-books. Additionally, with hundreds of thousands of interactions per day, ChatGPT is believed to be constantly improving its language model.
With players like Microsoft and Salesforce already looking to integrate ChatGPT capabilities into their workflows, it’s no wonder why demand for Generative AI is so high.
Given its popularity and potential, Inbenta is also often asked: How can businesses use Generative AI to achieve a business goal? What benefits can Generative AI bring to the customer experience more broadly?
Coding capabilities aside, the exact potential of Generative AI in a business context, used as a conversational tool, remains experimental. While Generative AI can help businesses instantly develop content and responses that can be used in a conversational environment, its client/customer readiness is questionable.
ChatGPT has the ability to chat, that is, to answer questions and facilitate conversations with users.
But the accuracy and relevance of ChatGPT responses can pose risks to businesses implementing ChatGPT, especially when brand voice and control are important.
To be fair, no large language model is perfect. When tested, ChatGPT has proven to have its own loopholes and instances of bias. Examples of inaccurate and odd conversations are well documented in press reports on the topic and can also be spotted after a few contacts.
There are also practical reasons why today’s version of ChatGPT is not ready for the conversational tasks required by businesses.
First, ChatGPT can be easily scaled. Enterprises can’t spend hundreds of hours tinkering with a language model to make sure it’s ready for their specific use case. Conversational AI tools need to be trained on business use cases and ready to use without a tab.
Second, the language model alone cannot sufficiently assist the user in performing personalized queries. For example, if a user wants to check their account balance or change purchase details, the language model must go beyond its original intent and be able to access the account information, authenticate the user, and accurately read and/or act on the request; These types of transactional integrations are widely seen in today’s conversational AI tools, including the Inbenta Chatbot, but not within ChatGPT.
Third, businesses should require their Generative AI responses to be validated by their marketing, sales, and legal teams. Off-topic and off-brand responses are not only a poor use of conversational AI, but should also be checked for authority. (No business wants its chatbot to engage in lewd or offensive exchanges.)
If you were to ask ChatGPT which business use cases it supports, you’d get a list of five capabilities: customer service and support; content creation and marketing; language translation; data analysis and insights; and use as a virtual assistant.
The reality is that today’s ChatGPT cannot actually deliver these use cases in a customer/customer ready manner without significant business intervention. Longer term, however, it’s easy to see how more accurate, tested, and connected Generative AI could be used to supercharge these tasks.
So far, conversational AI designed for these specific customer-centric use cases is being adopted by businesses around the world every day, with positive results. It is important that businesses apply AI responsibly. This includes approaching Generative AI with interest and appropriate caution.