A Reddit user shared what appears to be the system prompt for OpenAI's models, detailing their fundamental purpose, instruction hierarchy, and guidelines on honesty, truthfulness, objectivity, and neutrality. The prompt emphasizes that while user instructions can influence responses, they do not override higher-priority system-level requirements. It also highlights the model's expectation to represent information accurately, distinguish between facts and speculation, and avoid misleading implications, while acknowledging that generated information may not always be factually correct. AI
IMPACT Provides insight into the operational guidelines and safety protocols governing large language models.
RANK_REASON User-shared content that appears to be an internal policy document, not an official release.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
