The emergence of large language models (LLMs) such as ChatGPT has revolutionized information retrieval and interaction, particularly in intelligence analysis. However, concerns about their trustworthiness in security-critical applications persist due to security risks, factuality issues and biases. The project proposes a multifaceted approach to evaluating and enhancing the trustworthiness of large language models (LLMs), addressing key concerns related to security, factuality and biases.

The Multisource AI Scorecard Table (MAST) for Evaluating Generative AI in Worker-Automation Team Tasks (MEGAWATT) project aims to leverage MAST, a tool rooted in Intelligence Community Directive (ICD) 203, to evaluate and enhance the performance of large language models (LLMs) like OpenAI’s ChatGPT.

The CHIAUS project aims to revolutionize risk communication in software development and consumption by integrating human-centered interactions with Software Bill of Materials (SBOM) data. Led by Principal Investigator L Jean Camp and her team at Indiana University, the project addresses the pressing need for actionable, understandable risk communication in the software ecosystem.

In the realm of safety science, a paradigm shift towards resilience engineering offers new insights into addressing the challenges of complex systems, particularly in high-stakes environments like security screening. This project, Resilience Engineering for Visual Screening in Security Settings (REVS), aims to advance the understanding of resilience engineering principles within the context of security screening tasks.