Reputation: 1
I'm working through Azure OpenAI and a little confused about the privacy. The docs speak to the fact that nothing leaves the sanctity of the Azure environment. However, when I look at the code in the playground, I see the openai library being imported (see below). The note at the top seems to indicate it's a library specifically for Azure OpenAI, but that library name seems to be the generic openai library. So the question is "are all my prompts and completions really staying within Azure OpenAI like the docs indicate or am I sending private information to openai"?
#Note: The openai-python library support for Azure OpenAI is in preview.
import os
import openai
openai.api_type = "azure"
openai.api_base = "https://XXXXXXXX.openai.azure.com/"
openai.api_version = "2022-12-01"
openai.api_key = os.getenv("OPENAI_API_KEY")
response = openai.Completion.create(
engine="PRO-DI-OPENAI",
prompt="### Postgres SQL tables, with their properties:\n#\n# Employee(id, name, department_id)\n# Department(id, name, address)\n# Salary_Payments(id, employee_id, amount, date)\n#\n### A query to list the names of the departments which employed more than 10 employees in the last 3 months\n\nSELECT",
temperature=0,
max_tokens=150,
top_p=1,
frequency_penalty=0,
presence_penalty=0,
stop=["#",";"])
I've tried looking at the docs and general searches, but not seeing anything definitive.
Upvotes: 0
Views: 159
Reputation: 20117
If you use the generic OpenAI client with the Azure base api url endpoint as indicated, you're only sending data to Azure.
You can verify this by running wireshark or another network monitor to double-check where your traffic is going.
Upvotes: 0