We're engineers. We measure, we benchmark, we run postmortems. We pride ourselves on not taking things on faith. So why do we quote advice we've never traced to its source? Why do we copy architectures off conference talks, and pass "best practices" down to juniors without ever asking where they came from? Here's the uncomfortable part: most of the advice we repeat was genuinely true once — in a specific context that has quietly expired. The conclusion survived; the reasoning didn't. We inherited the answer without the question. This talk traces a handful of beliefs we all hold back to the moment they were actually true, watches that moment disappear, and asks what we should do instead. It starts somewhere you won't expect, and ends with a single question that protects you from the next expired truth — including the ones you're teaching right now.
AI Tech Lead @ Finonex