Is it safe to use text-davinci-003 as an endpoint in any application

API
1

Hi,
I am new to chatGPT and using text-davinci-003 to connect to the server. Is it safe to use ChatGPT in the enterprise application if I take precaution of not providing any sensitive information?
Also, the data sent to the server while using the API will be the data that I queried for, not any other data from the system?

Thanks in advance.

2

This is not possible to answer. “Safe” means different things to different people and organizations. If you can be more explicit with what you mean by “safe” perhaps we can help answer that for you.

OpenAI only receives what is sent. There is no mechanism by which OpenAI can access information on your computer without you including it in the API request payload.

1 Like
3
  1. You probably want to be looking at the GPT 3/4 models unless you are sure you want davinci
  2. Look into and understand the rate limits, data policy, billing limits to better understand if it’s viable for your use.
  3. The APIs can still be a bit flaky, plan for slow requests, occasional 429 errors, etc
  4. Also look into Azure OpenAI, may be more reliable