How to beginner · 3 min read

How to constrain AI behavior with system prompt

Q: How to constrain AI behavior with system prompt

Use the system prompt to set explicit instructions that constrain AI behavior before user input. This controls tone, style, and output scope by defining rules or roles, ensuring consistent and safe responses. For example, in OpenAI's chat.completions.create, pass a system message with constraints to guide the model.

Quick answer

Use the system prompt to set explicit instructions that constrain AI behavior before user input. This controls tone, style, and output scope by defining rules or roles, ensuring consistent and safe responses. For example, in OpenAI's chat.completions.create, pass a system message with constraints to guide the model.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the OpenAI Python SDK and set your API key as an environment variable to securely authenticate requests.

bash

pip install openai>=1.0

Step by step

Use the system message to constrain AI behavior by specifying clear instructions. Below is a complete example using the OpenAI SDK v1+ to create a chat completion where the AI is instructed to respond only with polite, concise answers.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a polite assistant who answers concisely and never provides opinions."},
        {"role": "user", "content": "Explain quantum computing in simple terms."}
    ]
)

print(response.choices[0].message.content)

output

Quantum computing uses quantum bits that can be in multiple states simultaneously, enabling faster problem-solving for certain tasks.

Common variations

You can constrain behavior with other models like Anthropic's Claude by using the system= parameter. Also, streaming responses or async calls can be used for real-time constraints.

python

import os
import anthropic

client = anthropic.Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

message = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=200,
    system="You are a helpful assistant who only answers factual questions and refuses to speculate.",
    messages=[{"role": "user", "content": "What is the meaning of life?"}]
)

print(message.content[0].text)

output

I'm here to provide factual information. The meaning of life is a philosophical question without a definitive factual answer.

Troubleshooting

If the AI ignores constraints, ensure your system prompt is explicit and placed as the first message. Avoid ambiguous language and test with different phrasings. Also, verify you are using the correct SDK version and model that supports system prompts.

✅

Key Takeaways

Always place the system prompt as the first message to set behavior constraints.
Use explicit, clear instructions in the system prompt to guide AI responses effectively.
Different SDKs require different ways to pass system instructions; follow their latest patterns.
Test constraints with varied inputs to ensure consistent AI behavior.
If constraints fail, refine prompt clarity and verify SDK and model compatibility.

Verified 2026-04 · gpt-4o, claude-3-5-sonnet-20241022

Verify ↗