video1.0<iframe src="https://www.loom.com/embed/77a63fb32c834deb8d18dd10b6e89faf" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/77a63fb32c834deb8d18dd10b6e89faf-5a8a90d179d6caed.gif204.502In this video, I demonstrate the audit process for AI agents, focusing on reasoning to prevent issues like hallucinations and FOMO. We ran several audits, starting with a FOMO strategy that resulted in a fail verdict, followed by another failure due to hallucinated data. I then presented a more structured strategy that successfully passed the audit. The results showed low to medium risk levels, and I encourage you to review the data and base scans provided. Please pay close attention to these findings as we refine our strategies.