Everyone does it, they all exfiltrate valuable data from OpenAI. You can use it directly, like Alpaca, or for pre-labelling, or for mislabeled example detection.

They train code models by asking GPT3 to explain code snippets, then training a model the other way around to generate code from description. This data can be used to fine-tune a code model for your specific domain of interest.