System Prompt
A system prompt is the hidden instruction that sets an AI assistant's behavior and rules before it sees the user's question, shaping how it answers and what it's allowed to do.
A system prompt is the assistant's standing instructions. Set by the product team before any conversation, it defines the model's role, tone, guardrails, and how it should use tools like search — context the user never sees but that governs every answer the assistant gives.
For AEO it's a reminder that you don't control the engine's behavior, only your content. The system prompt and the engine's design decide whether and how sources are cited, how cautious answers are, and what gets refused. Your leverage is making your page the clearest, most trustworthy input within whatever rules the engine imposes — aligned to the real question and easy to use. Trying to "trick" an assistant's instructions, by contrast, edges toward prompt injection and isn't a sustainable strategy.
Example. One engine's system prompt may tell it to always cite sources and hedge on medical topics, while another's prioritizes brevity. The same page can be cited differently across them — so you optimize the content, not the hidden instructions.
Relevant pillar
Related terms
- Prompt EngineeringPrompt engineering is the practice of crafting inputs to an AI model to get better, more reliable outputs, and in AEO it underlies how you build the prompt sets used to measure visibility.
- TemperatureTemperature is a setting that controls how random or deterministic an AI model's output is, with higher values producing more varied responses and lower values more predictable ones.
- Large Language Model (LLM)A large language model is an AI system trained on vast amounts of text to predict and generate language, and is the engine that writes the answers in AI search.