10. Mai 2024
The first test version of the BAUR ChatBot Kathi went online in December 2023. The latest version works on the basis of a Large Language Model (LLM) - a first for the Otto Group.
For the Customer Experience & Service team and Managing Director Stephan P. Elsner, there was no alternative to the idea of improving the shopping experience for our online store's customers by using generative AI.
In order to use this current state of technical development for baur.de, we contacted our colleagues from the Otto Group. Some of them were already working on similar solutions in parallel. Thanks to the collaboration with Otto Digital Consulting, it was possible to use this existing knowledge within the Group to develop our own ChatBot for baur.de.
The name Kathi was quickly found and the live test on the website was launched on December 1, 2023. The ChatBot is currently being used on the "Service and Help" page, where it answers our customers' questions about our services and advice.
The test phase, which was used to further develop Kathi, has now ended. The current version of the ChatBot works on the basis of a so-called Large Language Model (LLM) - a first for the Otto Group.
Special thanks go to Eva-Maria Nietner, who has driven the project forward from the very beginning with a great deal of commitment and expertise.
The aim of the development is to be able to expand the ChatBot even further in the future. It will also be available on other pages in the future and will be available to our customers there to make it easier for them to navigate the site.
Findings from Use Labs have shown that it can be a challenge for visitors to find the information they are looking for straight away. This is exactly where Kathi can come to the rescue in future and guide visitors to the right page.
For the answer, we asked the OTTO Group's generative AI, ogGPT:
A Large Language Model (LLM) is a powerful AI model that has been trained to understand and generate natural language. It is based on deep learning techniques and uses large amounts of text data to learn patterns and relationships in language.
An LLM consists of several layers of interconnected artificial neural networks. These models are capable of handling complex language tasks such as text comprehension, translation, text generation and even dialogs.
A well-known example of a Large Language Model is GPT-3 (Generative Pre-trained Transformer 3), which was developed by OpenAI. GPT-3 has 175 billion parameters and can demonstrate amazing linguistic capabilities by generating human-like text.
Large Language Models have the potential to be used in many areas, such as automatic translation, text generation, chatbots, content creation and even in supporting creative processes. However, there are also ethical issues surrounding the use of LLMs, as they could also be misused to spread misinformation or manipulative content. It is therefore important to use LLMs responsibly and assess their impact carefully.
If you would like to try out what Kathi can already do, you can find our Service & Help page here.
Die mit einem Sternchen (*) gekennzeichneten Felder sind Pflichtfelder. Bitte fülle diese aus, um deine Kommentare erfolgreich abgeben zu können.
Welcome to our BAUR Group JobsBlog. My name is Sylvain and as part of my internship in Corporate Communications at BAUR, I am currently supporting the HR Marketing team. Look forward to interesting blog posts and take a look behind the scenes of our company with me. 😊
Have fun reading!