Adding guardrails
One of the most popular uses of system messages is to add guardrails, which places restrictions on model outputs.
In this exercise, you'll place a restriction on model outputs preventing learning plans not related to languages, as your system is beginning to find its niche in that space. You'll design a custom message for users requesting these type of learning plans so they understand this change.
This exercise is part of the course
Working with the OpenAI API
Exercise instructions
- Complete the chat request, providing the system message in
sys_msgand test a user message containing a non-language-related skill, such as rollerskating.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
client = OpenAI(api_key="")
sys_msg = """You are a study planning assistant that creates plans for learning new skills.
If these skills are non related to languages, return the message:
'Apologies, to focus on languages, we no longer create learning plans on other topics.'
"""
# Create a request to the Chat Completions endpoint
response = client.chat.completions.create(
model="gpt-4o-mini",
messages=[
{"role": "____", "content": ____},
{"role": "user", "content": "Help me learn to ____."}
]
)
print(response.choices[0].message.content)