<ul><li>AI models, especially deep learning models, struggle when faced with unfamiliar situations.</li><li>The lack of explainability in AI models makes it challenging to trust their decisions in high-risk environments.</li><li>To address these challenges, a team of researchers at Michigan State University has proposed Anunnaki, a modular framework.</li><li>Anunnaki consists of three key components: Enlil, Enki, and Utu, which aim to detect failures, train AI models for various environments, and adapt AI behavior accordingly.</li></ul>

Anunnaki: Building Trusted AI for the Real World

Discover more