Episode 36: When Do You Trust AI Implicitly?

How often do you keep a close eye on AI? Mike's advice is to treat AI like a junior developer: set guardrails, test that it's working, and gradually give it more responsibility. The devs also talk about AI-generated FileMaker code, deterministic vs. non-deterministic workflows, prompt engineering, and where AI can be especially useful for automating repetitive tasks.There's also the idea of AI inside client applications, from document processing and data analysis to the potential for AI-powered reporting. Is 100% trust ever the goal? Or is AI most valuable as an advisor and productivity…