There’s no point in talking down to LLMs because they aren’t an entity that feels anything, and your aggressive words are just going to make it predict a response with those in the context. Polite requests get the best responses from humans so LLMs will be similar, not because they have feelings but because their training data likely includes both polite and aggressive communication and you’ll probably get better responses in the polite context.
There’s no point in talking down to LLMs because they aren’t an entity that feels anything, and your aggressive words are just going to make it predict a response with those in the context. Polite requests get the best responses from humans so LLMs will be similar, not because they have feelings but because their training data likely includes both polite and aggressive communication and you’ll probably get better responses in the polite context.