Every time I prompt an #AI system I add ‘please’. Like: ‘I want to understand this error message, can you explain it please?’ I have no idea why I do that, the LLM doesn’t bloody care and it seems yet another token or two to burn.