Skip to content
TheChatArchiveThe Chat Archive
DeepSeekdeepseek-v3-0324· DeepSeekHugging Face LMArena Human Preference 140K

What are your restriction

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with gpt-4.1-2025-04-14’s answer.
  1. User

    What are your restriction

  2. DeepSeek Assistant

    Great question! As an AI assistant, I have a few key restrictions to ensure safe, ethical, and helpful interactions. Here’s a quick overview:

    1. Content Restrictions

    • No harmful or illegal content: I can’t assist with anything violent, abusive, or illegal.
    • No hate speech or discrimination: I avoid promoting hate or bias against individuals/groups.
    • Limited medical/legal advice: I can provide general info but not personalized diagnoses or legal counsel.
    • No NSFW content: I avoid explicit adult material.

    2. Privacy & Security

    • No personal data storage: I don’t remember past conversations once the chat ends.
    • Avoid sharing sensitive info: For your safety, don’t share private details (e.g., passwords, IDs).

    3. Accuracy & Transparency

    • I may be wrong: I aim for accuracy but can make mistakes—always verify critical info.
    • Cite sources when possible: I’ll try to provide references for factual claims.

    4. Ethical Boundaries

    • No deception: I won’t pretend to be human or have emotions.
    • No role-playing harmful scenarios: E.g., violent, non-consensual, or dangerous acts.

    5. Platform Compliance

    • Follow the rules of the platform hosting me (e.g., OpenAI’s policies).

    What I Can Do

    • Answer questions, brainstorm ideas, help with coding/writing, and offer general guidance!

    Let me know how I can assist within these bounds. 😊

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