Generative AI provides an innovative and exciting way to manage knowledge and data at any scale; for small projects, at the enterprise level, and even at a world wide web scale. It is tempting to think that Generative AI has made other knowledge-based technologies obsolete; that anything we wanted to do with knowledge-based systems, Knowledge Graphs or even expert systems can instead be done with Generative AI. Our position is counter to that conclusion.Our practical experience on implementing enterprise question answering systems using Generative AI has shown that Knowledge Graphs support this infrastructure in multiple ways: they provide a formal framework to evaluate the validity of a query generated by an LLM, serve as a foundation for explaining results, and offer access to governed and trusted data. In this position paper, we share our experience, present industry needs, and outline the opportunities for future research contributions.

Collaborative development of a shared or standardized ontology presents unique issues in workflow, version control, testing, and quality control. These challenges are similar to challenges faced in large-scale collaborative software development. We have taken this idea as the basis of a collaborative ontology development platform based on familiar software tools, including Continuous Integration platforms, version control systems, testing platforms, and review workflows. We have implemented these using open-source versions of each of these tools, and packaged them into a full-service collaborative platform for collaborative ontology development. This platform has been used in the development of FIBO, the Financial Industry Business Ontology, an ongoing collaborative effort that has been developing and maintaining a set of ontologies for over a decade. The platform is open-source and is being used in other projects beyond FIBO. We hope to continue this trend and improve the state of practice of collaborative ontology design in many more industries.

In this paper, we present a framework for maintainable software design, based on a multilevel understanding of software function. This framework is the novel functional representation ZD, in which domain concepts (e.g., employee records) as well as computational concepts (e.g., stacks and queues) are represented in a reusable manner. We use ZD to support the construction of a library of reusable components for novel configurations. In order to achieve this, we provide mechanisms for supporting flexible configurations, address problems of determining the correctness of a combination of library components, and consider the computational complexity of finding combinations. These problems all stem from problems of interactions between components. Therefore, the structure of ZD is focused on representing and handling these interactions. We show how the approach based on ZD resolves certain well-known problems faced by other library-based component reuse architectures.

A current joint project between three institutions in Switzerland has as its goal to create Artificial Intelligence (AI) software in teaching principles of AI at the University level. The modules of this project, the Portable AI Lab (PAIL), illustrate basic concepts of Artificial Intelligence in a uniform and self-contained manner. This paper discusses the design considerations that were adopted in order to make the presentation of this material easier for novice students.

Although much of past work in AI has focused on compiled knowledge systems, recent research shows renewed interest and advanced efforts both in model-based reasoning and in the integration of this deep knowledge with compiled problem solving structures. Device-based reasoning can only be as good as the model used; if the needed knowledge, correct detail, or proper theoretical background is not accessible, performance deteriorates. Much of the work on model-based reasoning references the ‘no-function-in-structure’ principle, which was introduced by de Kleer and Brown. Although they were well motivated in establishing the guideline, this paper explores the applicability and workability of the concept as a universal principle for model representation. This paper first describes the principle, its intent and the concerns it addresses. It then questions the feasibility and the practicality of the principle as a universal guideline for model representation.