Have you ever given an artificial intelligence agent a clear instruction to avoid a specific action, only to watch it carry out that exact task anyway?

We suspect many of you have encountered this firsthand, and we would love to hear about your experiences. Please feel free to share your accounts in the response section below.