How to Ask ChatGPT to Stop Training on Your Content
I got a super juicy question for this month’s LinkedIn Live (I hold them the last Wednesday of every month a noon ET 😊):
How to use AI to help create a curriculum with my own IP without permitting AI to use that IP when responding to prompts from others?
So, of course, I had to ask ChatGPT. First, I got what I expected blah blah blah:
“Use of Content: OpenAI may use the content you provide to maintain, develop, and improve its services. This includes training their models, enforcing terms and policies, and ensuring the safety of the services.”
But then I got something unexpected:
If you prefer not to have your data used for model training, there is an option to opt-out, although this may affect the service’s ability to cater specifically to your needs. [emphasis added]
Well, that’s interesting. Instructions are here.
Of course, this option is not without its downside. I’m sure you noticed that turning off model training and turning off chat history go hand in hand. I would be loath to lose chat history.
So there are costs and benefits to be weighed. Personally, I will keep full function paired with some best practices:
Never include sensitive, confidential or proprietary information in prompts.
Always remove personally identifiable information, such as names and contact details.
Be aware of any contractual restrictions that prohibit use of AI in connection with client (and other third party’s) materials.
Would love to hear your thoughts – To train or not to train?
Cheers! 🥂
Erin