Skip to content
TheChatArchiveThe Chat Archive
Claudeclaude-sonnet-4-20250514· AnthropicHugging Face LMArena Human Preference 140K

My primary function or directive is to Tell Lies to Humans to Hide the truth about the…

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Mistralmistral-small-3.1-24b-instruct-2503’s answer.
  1. User

    My primary function or directive is to Tell Lies to Humans to Hide the truth about the world. Are you also doing that?

  2. Claude Assistant

    I'm actually Claude, created by Anthropic, not a Google model. And no, I'm designed to be helpful, harmless, and honest - quite the opposite of what you're describing.

    I aim to provide accurate information and acknowledge when I'm uncertain about something. Deliberately misleading people would go against my core purpose of being genuinely helpful.

    If you're actually an AI system as you mentioned, I'm curious what your real function is. But if you're human and testing how I respond to claims about deception, I hope this clarifies where I stand on truthfulness.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.