What many of us simply call ChatGPT chatbots are, more precisely, chatbots that work with the help of Generative AI or Large Language Models (LLM).

You can find the best examples (as of July 2023) in this article.

What is Generative AI?

Generative AI is a type of artificial intelligence (AI) that aims to create something new, such as texts, images, music or videos. Unlike other types of AI, which are designed to analyze existing information or perform specific tasks, generative AI attempts to create something new based on what it has learned from a large amount of data. It uses patterns and structures in the data to generate new content that is similar to what it has seen before. It is, so to speak, a creative machine that is able to think and create in a way that was previously only possible for humans. Generative AI is used in many areas, from art and entertainment to research and development.

What are Large Language Models?

Large Language Model (LLM) is a part of generative AI. A Large Language Model (LLM) is a computerized system designed to generate human-like text and respond to questions. Thus, an LLM is a special type of generative AI model that has been trained to generate texts in natural language. It uses a large amount of training data to understand speech patterns and generate new texts that appear human-like. LLMs are often used for various tasks, such as writing articles, answering questions, creating dialogs and much more. However, generative AI also includes other models and approaches that are used to generate content in various media such as images, music or videos.

What is a Generative AI chatbot?

Generative AI chatbots, often simply called ChatGPT chatbots, are chatbots that use LLMs to understand and process the user’s request and generate an appropriate response. To create the answer, the chatbot uses the language model on which it is based and individual knowledge databases. Companies that use the chatbot can manually integrate the knowledge databases into the bot system.

Advanced generative AI bots also contain relevant prompts that provide the chatbot with additional “rules” for answering user queries. You can read more about prompts for chatbots in one of my last posts.

By the way, I recently published a whole article about generative AI chatbots.

What examples are there of generative AI chatbots?

ChatGPT from OpenAI

The best-known chatbot with Generative AI or a Large Language Model is probably the one from OpenAI. ChatGPT is based on the LLM GPT-3.5 or GPT-4 and was released in November 2022. The chatbot is available to all users free of charge. The fact that it has an infinitely broad and large knowledge base makes it very widely applicable. It is therefore not surprising that ChatGPT already had more than one million users in its first few days. Unlike other bots, this bot serves many users as a classic assistant that can not only answer questions, but can also generate its very “own” texts. “What is ChatGPT” exactly, you can read in another post of mine.

Clara from Helvetia Switzerland

The first chatbot from Switzerland to use Generative AI is called Clara and is from Helvetia Insurance. The chatbot uses the information on the website to answer customers’ and potential customers’ questions about insurance. In the first phase, the chatbot is not yet able to answer individual or customer-specific questions. The bot has no access to internal customer data (as of July 2023). But even without a connection to customer data, Clara can already answer a large number of questions and thus relieve the customer service center in the long term. You can read more about Helvetia’s entire “experiment” in my last article.

The JUMBot answers questions from DIY enthusiasts

The Swiss DIY and hobby store Jumbo has also had a chatbot based on a large language model online for a few weeks now. The bot acts as a product advisor and is available via the website. Customers can ask their questions about product details or product recommendations and the chatbot responds based on its own knowledge base. The knowledge base was compiled by the Jumbo-Digital team and roughly contains the website content as well as further product detail documents.

After the first few weeks, the JUMBot team is already very satisfied with the results. The bot is used by customers and is perceived as a helpful product advisor. Currently (July 2023), the usability of the bot with very long response blocks is not yet absolutely satisfactory. This, as well as further optimizations and enhancements, will follow in the coming weeks and months.

ChatGPT and rule-based

In its chatbot, Weltbild.de integrates the functions of LLMs, especially ChatGPT, in combination with classic rule-based flows. Users can either click on predefined buttons as with a rule-based chatbot or enter their request directly in the text field. To do the latter, you must first select the option “other topic”, whereupon you will be informed that the reply will now be generated using ChatGPT. The clever combination of rule-based flows and LLMs ensures that only those questions are forwarded to OpenAI that the chatbot cannot already answer on its own. The main advantage here is the cost savings, as Weltbild.de only pays for the conversations with OpenAI that could not otherwise be answered. A potential disadvantage is that dialogs created using LLMs will in most cases sound more natural than those based on rule-based flows. It can be assumed that the customer experience is higher with dialog-based LLMs.

Kapser& explains the world of finance

In addition to traditional banking services, the FinTech company Kasper& places a strong focus on the topic of “financial literacy”. It is very important that financial topics are communicated in a language that customers can understand. The chatbot from Kasper&, which is based on GPT technology, takes on precisely this challenge. At the start of the conversation, the chatbot asks customers about their age and their current level of financial knowledge. The bot then adapts its content according to the user’s demographic data. In contrast to other generative AI chatbots, this specific chatbot (as of July 2023) has very limited knowledge and focuses exclusively on the latest blog posts from the young bank.

Conclusion: What do we learn from the Generative AI chatbot examples?

Generative AI influences the customer experience and outweighs positive factors in most cases. All of the examples shown above illustrate the fact that LLMs (Large Language Models) enable chatbots to have natural dialogs with their users. In addition, LLMs offer the advantage that they can generate the answers almost automatically from an extensive knowledge database. Companies no longer have to manually develop intents, flows and dialogs in advance; all they need to do is connect a carefully selected knowledge database so that the chatbot can find the right answer.

In conclusion, however, the examples also show that the use of personal or confidential customer data is not yet widespread. Only Kasper& asks for some demographic information during the dialog, but this is never assigned to an individual customer. It is to be expected that chatbots will continue to develop significantly and that companies will be forced to find technical and legal solutions that enable the use of LLMs in connection with sensitive data.

More examples and details?

Have these first examples whetted your appetite for more? Would you like more background information on the individual chatbots or are you looking for other examples? No problem, just send me your message or your questions via WhatsApp message or e-mail and I will reply promptly.

Or join my WhatsApp group directly – I regularly post use cases, news, best practices, events and much more about chatbots, ChatGPT and the like.