Menu
Menu
Since the end of 2022, Generative AI has been a widely discussed topic. After the initial amazement at the capabilities of ChatGPT, surpassing expectations from a few years ago, we are now clearly at the point where many companies in various industries want to harness this cutting-edge technology.
Does it make sense? In our opinion, definitely yes. We should not close ourselves off to new technologies or cling to the hope that a new technology cannot perform something as well as the existing solution. At the same time, we should not naively make use of every innovation. Prior to use, two questions are central:
Especially regarding the second point, there are interesting insights, both in terms of data security and hacker attacks. Our colleagues from sequire technology extensively reported on this last year and published a paper on the topic of “Indirect Prompt Injection”.
But back to the original question: Where can players in the automotive aftermarket appropriately apply Generative AI? Our COO Kevin Dewi presented a very illustrative example at this year’s DIGITAL COMMERCE SUMMIT automotive & industry:
The automotive aftermarket is a highly complex industry. Finding the right replacement parts, such as brake disks, for a specific vehicle model is not a simple Task. Additional questions arise. Do I need to replace the other wear parts beside the brake disks themselves? How do I best conduct the repair? What specific considerations should I keep in mind? Our platform N4PARTS provides competent assistance in addressing these challenges.
Retrieval Augmented Generation (RAG) is an innovative approach in artificial intelligence based on the combination of retrieved information and generated content. In this model, relevant information is first retrieved from large text datasets (‘retrieve’). Subsequently, this information is used to guide and enhance the generation of new, contextually relevant texts (‘generate’). By combining retrieval techniques with generative models, RAG enables precise and rich text generation based on a broad knowledge foundation. This approach is applied in various areas such as question-answering systems, text processing and content creation to produce high quality and relevant texts that align with the requirements and context. RAG thus represents and advancement in the utilization of the text data and generation to develop powerful and adaptive AI systems.
Our approach? An assistant, in the form of a co-pilot, that makes the dialogue between humans and machines more intuitive. This makes the tool easier to use, and we can provide added value by offering supplementary information that wasn’t directly requested but is relevant for the task at hand. Additionally, using Retrieval Augmented Generation (RAG) , we ensure that the language model has access to the right documents to answer the question, making the response valid and transparent to the user by displaying the sources at all times—an additional layer of security so that users don’t solely rely on the statements of a language model.
DR. CHRISTOPH ENDRES
CEO
sequire technology
Other articles that might be interesting for you
