It's legal if I do not tell my user that they are chatting with Gpt-model but in another name (eg. ChatBot-X, ChatBot-XL)

Question

I'm using Azure OpenAI API for bot conversation and integrated into our Platform's features. It's legal if I do not tell my user that they are chatting with Gpt-model but in another name (eg. ChatBot-X, ChatBot-XL)

Answer 1

@Hoang Truong Firstly, I am not a lawyer, so this should not constitute legal advice and you should always check with a lawyer for legal requirements which may vary by region and industry.

Again, just to be perfectly clear, this would be completely dependent on the region where you are operating and/or the industry you are building for.

In general, it would be best for both your users and you to let them know they are talking to an AI System rather than a human, simply to set the right expectations, especially considering that you can't always predict how a generative model will behave.

This is exactly why the Responsible AI Practices were introduced to help everyone build upon and use AI in their products.

There may be certain cases where AI is used but users do not interact directly with like AI-powered Spam Filters for example, which may not necessarily warrant a disclaimer considering it is part of the system that they are not privy to. But then again, it would be best to check local regulations to be sure. Better safe than sorry. 🙂