This is the first BTS ByteSized episode. Shorter discussions on offshore life, diving, inspection, NDT and where the industry is heading. If you are a premium member you can listen earlier than everyone else as its in the Pre-Dive Podcast section right now otherwise it'll be out next week on Spotify and wherever else you listen to your podcasts. People are judging AI in inspection using the wrong tool for the wrong task. After seeing a LinkedIn post where ChatGPT failed to correctly interpret a DPI indication, I wanted to discuss what that actually means for the future of inspection and NDT. In this episode I break down: The difference between large language models and computer vision AI Why pattern recognition is central to inspection How AI could assist techniques like ACFM, MPI, DPI and UT Where AI may genuinely improve consistency and anomaly detection The limitations of AI offshore Why experienced human judgement still matters