video1.0<iframe src="https://www.loom.com/embed/62fa0ac8836b46baa640e8fe474cd6c3" frameborder="0" width="1728" height="1296" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12961728Loomhttps://www.loom.com12961728https://cdn.loom.com/sessions/thumbnails/62fa0ac8836b46baa640e8fe474cd6c3-3756edde9b0f58ed.gif141.691In this video, I shared our journey from developing an analytical AI for an insurance company to creating a robust system focused on safety, energy efficiency, and auditability. We now have over 65 agents across 6 roles and 39 knowledge domains that filter out malicious prompts, operating entirely on NPU CPU to save 70% of tokens. This innovation could prevent the need for 600 atomic reactors by 2045 and save approximately 14 trillion US dollars. Our goal is to achieve 12 Sigma security, as current AI safety levels could lead to catastrophic outcomes if applied to critical areas like aviation. I encourage you to explore our knowledge domains further by clicking on them to see how our prompt filtering works.